WO2023279371A1 - 自动驾驶方法、装置和存储介质 - Google Patents

自动驾驶方法、装置和存储介质 Download PDF

Info

Publication number
WO2023279371A1
WO2023279371A1 PCT/CN2021/105476 CN2021105476W WO2023279371A1 WO 2023279371 A1 WO2023279371 A1 WO 2023279371A1 CN 2021105476 W CN2021105476 W CN 2021105476W WO 2023279371 A1 WO2023279371 A1 WO 2023279371A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
value
road surface
point cloud
cloud data
Prior art date
Application number
PCT/CN2021/105476
Other languages
English (en)
French (fr)
Inventor
卢远志
赵世杰
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN202180006112.6A priority Critical patent/CN114616158A/zh
Priority to PCT/CN2021/105476 priority patent/WO2023279371A1/zh
Publication of WO2023279371A1 publication Critical patent/WO2023279371A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B60W60/0015—Planning or execution of driving tasks specially adapted for safety
    • B60W60/0016—Planning or execution of driving tasks specially adapted for safety of the vehicle or its occupants
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W10/00—Conjoint control of vehicle sub-units of different type or different function
    • B60W10/22—Conjoint control of vehicle sub-units of different type or different function including control of suspension systems
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/18—Propelling the vehicle
    • B60W30/18009—Propelling the vehicle related to particular drive situations
    • B60W30/18163—Lane change; Overtaking manoeuvres
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08—Interaction between the driver and the control system
    • B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/50—Barriers
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2720/00—Output or target parameters relating to overall vehicle dynamics
    • B60W2720/10—Longitudinal speed

Definitions

  • the present application relates to the technical field of automatic driving, and in particular to an automatic driving method, device and storage medium.
  • an embodiment of the present application provides an automatic driving method, the method includes: acquiring data collected by sensors, the data including point cloud data in front of the vehicle and acceleration data of the vehicle in a direction perpendicular to the road surface; according to The point cloud data determines a feature value, and the feature value is used to indicate the undulation degree of the road ahead of the vehicle; at least according to the acceleration data and/or the feature value, an automatic driving strategy of the vehicle is determined.
  • the current unevenness of the road surface can be accurately identified.
  • the above eigenvalues are used to determine the automatic driving strategy of the vehicle, which can reduce the impact of unexpected steering caused by uneven roads on the precise control of the vehicle, improve the safety and comfort of the vehicle during automatic driving, and at the same time reduce the , Vehicle sensors and other structural damage due to vibration, thereby improving its reliability and life durability.
  • the feature value is determined according to the coordinate value of the point cloud data in a direction perpendicular to the road surface.
  • the coordinate value of the point cloud data in the direction perpendicular to the road surface can reflect the degree of undulation of the road surface. According to the embodiment of the present application, the characteristic value can better measure the unevenness of the road surface.
  • determining the feature value according to the point cloud data includes: determining the first feature value and the second feature value according to the point cloud data and at least one of the third eigenvalues, the first eigenvalue is used to indicate the undulation degree of the local area in each lane in the road ahead of the vehicle, and the second eigenvalue is used to indicate the road ahead of the vehicle Among them, the degree of undulation of the road surface of each lane, the third characteristic value is used to indicate the degree of undulation of the road surface in front of the vehicle, and the road surface of all lanes.
  • the first eigenvalue, the second eigenvalue, and the third eigenvalue by determining at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue, and the first eigenvalue, the second eigenvalue, and the third eigenvalue are obtained from points, lines, and surfaces respectively
  • the three angles indicate the degree of undulation of the road surface, which can more accurately and comprehensively reflect the unevenness of the road surface, thus providing a targeted basis for the subsequent determination of automatic driving strategies.
  • the automatic driving method at least according to the acceleration data and/or the characteristic value, determine the automatic driving of the vehicle
  • the strategy includes: according to the second eigenvalue, determining a target lane corresponding to the smallest second eigenvalue; and determining a strategy for the vehicle to switch to the target lane.
  • the strategy for the vehicle to switch to the target lane is determined, so that the vehicle can avoid the lane with a large degree of undulation to avoid uneven road conditions, Thereby improving the safety and comfort of the vehicle and reducing related structural damage.
  • the The automatic driving strategy of the vehicle includes: according to the first eigenvalue, determining a target local area corresponding to the first eigenvalue exceeding a threshold; and determining a strategy for the vehicle to avoid the target local area.
  • the vehicle avoidance target local strategy by determining the target local area corresponding to the first eigenvalue exceeding the threshold according to the first eigenvalue, the basis for determining the vehicle avoidance target local strategy, so that the vehicle can avoid local unevenness such as potholes in the driveway road surface to avoid uneven road conditions, thereby improving vehicle safety and comfort and reducing related structural damage.
  • determining the automatic driving strategy of the vehicle including: determining a strategy for limiting the driving speed of the vehicle according to at least one of the first characteristic value, the second characteristic value, and the third characteristic value, wherein the restricted driving of the vehicle
  • the upper limit value of the speed is negatively correlated with the maximum value of at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue.
  • the vehicle by determining a strategy for limiting the vehicle speed according to at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue, the vehicle can be decelerated before encountering an uneven road,
  • the upper limit value of the limited driving speed of the vehicle negatively correlated with the maximum value of at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue when determining the strategy for limiting the vehicle speed. It can be realized that the upper limit of the speed limit of the vehicle is smaller when the degree of road surface undulation is greater, so that the vibration amplitude of the vehicle can be further reduced, the safety and comfort of the vehicle can be improved, and related structural damage can be reduced.
  • determining the automatic driving strategy of the vehicle includes: when the characteristic value exceeds a threshold value, or the rate of change of the acceleration data exceeds a threshold value, determining the A strategy for tuning the parameters of the vehicle's active suspension.
  • the vehicle by determining a strategy for instructing to adjust the parameters of the active suspension of the vehicle when the characteristic value exceeds the threshold value or the rate of change of the acceleration data exceeds the threshold value, it can be realized that the vehicle passes through the uneven road. , adjust the parameters of the active suspension in advance, so as to reduce the vibration amplitude of the vehicle when passing the uneven road, improve the safety and comfort of the vehicle, and reduce related structural damage.
  • determining the feature value according to the point cloud data includes: when the rate of change of the acceleration data exceeds a predetermined threshold, determining The eigenvalues are adjusted to obtain the adjusted eigenvalues, and the adjusted eigenvalues are greater than the pre-adjusted eigenvalues.
  • the adjusted eigenvalues can reflect the current unevenness of the road more truly, so that the adjusted eigenvalues can be more accurate.
  • the eigenvalues generate a more accurate automatic driving strategy, thereby further improving the safety and comfort of the vehicle during automatic driving, and reducing related structural damage.
  • determining the feature value according to the point cloud data includes: dividing the region of interest in the road ahead of the vehicle into grids; deleting the point cloud data in the region of interest, corresponding to obstacles The point cloud data is obtained to obtain the processed point cloud data; according to the processed point cloud data in at least one grid, the feature value is determined.
  • the eigenvalues can reflect the unevenness of the road surface more accurately, so that better Auxiliary generation of relevant automatic driving policies.
  • determining feature values according to the point cloud data includes: according to one or one The coordinate values of the point cloud data in the group grid in the direction perpendicular to the road surface, determine the eigenvalues of the local area corresponding to the one or a set of grid data, and obtain the first eigenvalue; and/or according to any The coordinate values of the point cloud data in one or more columns of grids in the lane direction in the direction perpendicular to the road surface, determine the eigenvalue corresponding to any lane, and obtain the second eigenvalue; and/or according to all grids
  • the coordinate values of the point cloud data in the grid in the direction perpendicular to the road surface are used to determine the third feature value.
  • the angle of the surface measures the unevenness of the road surface.
  • the feature value is based on the z value of the highest point in the processed point cloud data, the z value of the lowest point, the difference between the z value of the highest point and the lowest point, z One or more of the average value of the value and the variance of the z value, the z value represents the coordinate value of the point cloud data in the direction perpendicular to the road surface.
  • the embodiment of the present application it is possible to flexibly select the method of determining the characteristic value according to the needs, so that the degree of road surface undulation indicated by the characteristic value can be more in line with the actual situation of the vehicle during automatic driving.
  • the method further includes: according to the strategy, generating an automatic driving strategy for the undulation degree of the road ahead of the vehicle and/or the vehicle
  • the prompt information includes at least one of text, image, audio, and video.
  • the driver and passengers by generating the prompt information, the driver and passengers can know the current road conditions and the relevant strategies adopted by the vehicle in a timely manner, thereby improving the trust of the drivers and passengers.
  • an embodiment of the present application provides an automatic driving device, which includes: an acquisition module, configured to acquire data collected by sensors, the data including point cloud data in front of the vehicle and the vehicle's direction perpendicular to the road surface The acceleration data; the first determination module is used to determine the feature value according to the point cloud data, and the feature value is used to indicate the undulation degree of the road surface in front of the vehicle; the second determination module is used to at least based on the acceleration data and/or or the feature value, to determine the automatic driving strategy of the vehicle.
  • the feature value is determined according to the coordinate value of the point cloud data in a direction perpendicular to the road surface.
  • determining the feature value according to the point cloud data includes: determining the first feature value and the second feature value according to the point cloud data and at least one of the third eigenvalues, the first eigenvalue is used to indicate the undulation degree of the local area in each lane in the road ahead of the vehicle, and the second eigenvalue is used to indicate the road ahead of the vehicle Among them, the degree of undulation of the road surface of each lane, the third characteristic value is used to indicate the degree of undulation of the road surface in front of the vehicle, and the road surface of all lanes.
  • the automatic driving device determines the automatic driving of the vehicle
  • the strategy includes: according to the second eigenvalue, determining a target lane corresponding to the smallest second eigenvalue; and determining a strategy for the vehicle to switch to the target lane.
  • the The automatic driving strategy of the vehicle includes: according to the first eigenvalue, determining a target local area corresponding to the first eigenvalue exceeding a threshold; and determining a strategy for the vehicle to avoid the target local area.
  • determining the automatic driving strategy of the vehicle including: determining a strategy for limiting the driving speed of the vehicle according to at least one of the first characteristic value, the second characteristic value, and the third characteristic value, wherein the restricted driving of the vehicle
  • the upper limit value of the speed is negatively correlated with the maximum value of at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue.
  • determining the automatic driving strategy of the vehicle includes: when the characteristic value exceeds a threshold value, or the rate of change of the acceleration data exceeds a threshold value, determining the A strategy for tuning the parameters of the vehicle's active suspension.
  • determining the feature value according to the point cloud data includes: when the rate of change of the acceleration data exceeds a predetermined threshold, determining The eigenvalues are adjusted to obtain the adjusted eigenvalues, and the adjusted eigenvalues are greater than the pre-adjusted eigenvalues.
  • determining the feature value according to the point cloud data includes: dividing the region of interest in the road ahead of the vehicle into grids; deleting the point cloud data in the region of interest, corresponding to obstacles The point cloud data is obtained to obtain the processed point cloud data; according to the processed point cloud data in at least one grid, the feature value is determined.
  • determining feature values according to the point cloud data includes: according to one or one The coordinate values of the point cloud data in the group grid in the direction perpendicular to the road surface, determine the eigenvalues of the local area corresponding to the one or a set of grid data, and obtain the first eigenvalue; and/or according to any The coordinate values of the point cloud data in one or more columns of grids in the lane direction in the direction perpendicular to the road surface, determine the eigenvalue corresponding to any lane, and obtain the second eigenvalue; and/or according to all grids
  • the coordinate values of the point cloud data in the grid in the direction perpendicular to the road surface are used to determine the third feature value.
  • the feature value is based on the z value of the highest point, the z value of the lowest point, the difference between the z value of the highest point and the lowest point, z One or more of the average value of the value and the variance of the z value, the z value represents the coordinate value of the point cloud data in the direction perpendicular to the road surface.
  • the device further includes: a generating module, configured to generate the fluctuation degree of the road ahead of the vehicle and/or the The prompt information of the automatic driving strategy of the vehicle, the prompt information includes at least one of text, image, audio, and video.
  • an embodiment of the present application provides an automatic driving device, which includes: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions
  • An automatic driving method that implements the first aspect or one or more of the multiple possible implementations of the first aspect.
  • the embodiments of the present application provide a non-volatile computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above-mentioned first aspect or the first aspect can be realized One or more of the various possible implementations of the automatic driving method.
  • the embodiments of the present application provide a terminal device, which can execute the automatic driving method of the above-mentioned first aspect or one or several possible implementation manners of the first aspect.
  • the embodiments of the present application provide a computer program product, including computer readable codes, or a non-volatile computer readable storage medium bearing computer readable codes, when the computer readable codes are stored in an electronic
  • the processor in the electronic device executes the automatic driving method of the first aspect or one or more of the multiple possible implementations of the first aspect.
  • Fig. 1 shows a schematic diagram of an application scenario according to an embodiment of the present application.
  • Fig. 2 shows a schematic structural diagram of an automatic driving device according to an embodiment of the present application.
  • Fig. 3 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • Fig. 4 shows a schematic diagram of determining features of points, lines and planes according to an embodiment of the present application.
  • Fig. 5 shows a schematic diagram of determining an automatic driving strategy according to an embodiment of the present application.
  • Fig. 6 shows a schematic diagram of visual icon prompting on a human-machine interface HMI according to an embodiment of the present application.
  • Fig. 7 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • Fig. 8 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • Fig. 9 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • Fig. 10 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • Fig. 11 shows a structural diagram of an automatic driving device according to an embodiment of the present application.
  • Fig. 12 shows a structural diagram of an automatic driving device according to an embodiment of the present application.
  • Fig. 1 shows a schematic diagram of an application scenario according to an embodiment of the present application.
  • the method of an embodiment of the present application can be used in the process of automatic driving of the vehicle on the road, where the vehicle may encounter uneven road conditions, as shown in Figure 1, the uneven road conditions may include bumps The road surface of the manhole cover, the road surface paved with speed bumps, and the potholed road surface.
  • the automatic driving method of the embodiment of the present application can use sensors to detect the characteristics of the uneven road surface.
  • the three vertical and vertical angles respectively generate relevant strategies, and realize the automatic driving of the vehicle according to the relevant strategies, so as to realize the vehicle passing through the uneven road surface, thereby reducing or avoiding the impact of the uneven road surface on the vehicle, so as to improve the vehicle's automatic driving process.
  • Safety and comfort while reducing the structural fatigue damage of sensors and vehicles, and improving their reliability and stability.
  • the situation of uneven road surface shown in FIG. 1 is only an example, and the situation of uneven road surface may also include other situations such as bumpy and undulating road surfaces, damaged road surfaces, gravel roads, and jointed road surfaces.
  • the self-driving approach can also be used in other scenarios.
  • Fig. 2 shows a schematic structural diagram of an automatic driving device according to an embodiment of the present application.
  • the automatic driving device can include a vehicle, and the vehicle can be a vehicle with a wireless communication function, wherein the wireless communication function can be set on the vehicle-mounted terminal, vehicle-mounted module, vehicle-mounted unit, chip (system) or other parts or components.
  • the vehicle in the embodiment of the present application may be in an automatic driving state, that is, the vehicle is completely autonomously driven without the driver's control or with only a small amount of driver's control.
  • At least one sensor can be set on the vehicle, such as vehicle radar (such as millimeter wave radar, laser radar, ultrasonic radar, etc.), rain sensor, camera, vehicle attitude sensor (such as gyroscope), inertial measurement unit (inertial measurement unit, IMU) , global navigation satellite system (global navigation satellite system, GNSS), etc., other sensors can also be set on the vehicle, and the above-mentioned sensors can be set on one vehicle, or on multiple vehicles, through at least one sensor set on the vehicle , the point cloud data of the road surface and the vertical acceleration of the vehicle (that is, the acceleration data of the vehicle in the direction perpendicular to the road surface) and other data can be collected.
  • vehicle radar such as millimeter wave radar, laser radar, ultrasonic radar, etc.
  • rain sensor such as gyroscope
  • IMU inertial measurement unit
  • global navigation satellite system global navigation satellite system
  • other sensors can also be set on the vehicle, and the above-mentioned sensors can be set on one vehicle,
  • the vehicle can also be equipped with an automatic driving system.
  • the automatic driving system can be used to generate an automatic driving strategy for dealing with uneven road conditions based on the data collected by the sensor, and realize the automatic driving of the vehicle according to the generated strategy, so as to enable the vehicle to pass through the uneven road surface. .
  • the vehicle can also be equipped with a human machine interface (human machine interface, HMI).
  • HMI human machine interface
  • the HMI can be used to broadcast the current road conditions and the strategies adopted by the automatic driving system for the vehicle through visual icons and voice broadcasts, so as to remind relevant drivers and passengers.
  • the automatic driving device of the present application may also include a server.
  • the server may be located on the above-mentioned vehicle as a vehicle-mounted computing unit, or may be located in the cloud, and may be a physical device or a virtual device such as a virtual machine,
  • the container and the like have a wireless communication function, wherein the wireless communication function can be set on the chip (system) or other parts or components of the server.
  • the server and the vehicle can communicate through wireless connections, such as mobile communication technologies such as 2G/3G/4G/5G, as well as wireless communication technologies such as Wi-Fi, Bluetooth, frequency modulation (FM), digital radio, and satellite communications.
  • the server can be carried on the vehicle and communicate with the vehicle through a wireless connection. Through the communication between the vehicle and the server, the server can collect The data collected by sensors on the road or other places is calculated, and the calculation results are sent back to the corresponding vehicles.
  • Fig. 3 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • the flow of the automatic driving method may include:
  • step S301 the vehicle acquires data collected by sensors.
  • the data collected by the sensor may include the point cloud data collected by the lidar and the vertical acceleration data of the vehicle collected by the IMU.
  • the point cloud data can be the point cloud data in front of the vehicle during the driving process of the vehicle.
  • the vertical acceleration of the vehicle can refer to the acceleration of the vehicle in the direction perpendicular to the road surface.
  • the vertical acceleration of the vehicle can be used to represent the vehicle’s speed on a certain road degree of vertical vibration.
  • the sensors can be installed on one or more vehicles, and the data collected by the sensors can be obtained by one vehicle.
  • the sensor can be set on a vehicle, and the sensor can also be set on other vehicles except the one vehicle to assist the automatic driving of the one vehicle.
  • the point cloud data collected by the vehicle is supplemented.
  • the sensor can also be arranged in other places, for example, the sensor can be arranged on both sides of the road, and can be used to collect point cloud data to supplement the point cloud data collected by the sensor on the vehicle.
  • step S302 the automatic driving system determines the point, line, and surface features of the road surface according to the data collected by the sensor.
  • the point, line, and surface features of the road surface can respectively represent the undulation degree of the road surface from three feature dimensions.
  • Fig. 4 shows a schematic diagram of determining features of points, lines and planes according to an embodiment of the present application.
  • the automatic driving system can first perform coordinate transformation on the obtained point cloud data to obtain the coordinates of each point in the point cloud data in the Cartesian coordinate system
  • x, y, and z can respectively represent the coordinate values of the corresponding point on the x-axis, y-axis, and z-axis, such as , in Figure 4
  • P i , P i+1 and P i+2 can respectively represent three points in the point cloud data in the Cartesian coordinate system.
  • the reference reference road surface can be determined by filtering the point cloud data, which can represent the horizontal reference surface of the road surface, and the more each point in the point cloud data deviates from the reference reference road surface in the z-axis direction, it can represent the corresponding The greater the undulation of the road surface, it should be noted that the reference reference road surface can also be determined in other ways.
  • the point cloud data of the road surface of the region of interest can be screened from the point cloud data.
  • the region of interest can include the driving road surface in front of the vehicle (that is, the lane in front of the vehicle), and the road surface, stationary or moving of the region of interest can be filtered out.
  • the point cloud data corresponding to the obstacle is the driving road surface in front of the vehicle (that is, the lane in front of the vehicle), and the road surface, stationary or moving of the region of interest can be filtered out.
  • the obtained point cloud data can be fitted by random sample consensus (RANSAC) to determine the lane, so as to determine the boundary of the region of interest, and the range of the region of interest can also be pre-set, for example , the ROI can take the leftmost lane line and the rightmost lane line in the driving direction as the left and right boundaries, and use the front of the vehicle and a preset distance (for example, 30 meters) in front of the vehicle as the front and rear boundaries.
  • the area of interest may include all lanes in the driving direction, or some lanes (for example, lanes other than emergency lanes, or lanes other than forbidden lanes). Obstacles may include other vehicles, pedestrians, cones, traffic lights, etc. on the road.
  • the value of the z-axis is outside the predetermined threshold range (for example, outside 50cm).
  • the area acts as an obstacle that needs to be screened out.
  • the automatic driving system can also perform grid processing on the area of interest in front of the vehicle.
  • the resolution of the grid can be fixed.
  • the size of the area of interest can be 30m*10m, and the resolution of each grid is 1m* 1m, the region of interest can be divided into 300 grids of the same size; the resolution of the grid can also be changed, for example, in the x-axis direction, as the distance between the road surface and the vehicle is farther, the corresponding point
  • the sparser the cloud, the grid resolution size can gradually increase from 1m*1m, for example, gradually change to 2m*1m.
  • the eigenvalue corresponding to each grid can be determined, and the eigenvalue of each grid can be, for example, the z value of the highest point in the point cloud data in the grid (that is, the z value on the z axis value), the z value of the lowest point, the z value difference between the highest point and the lowest point, the average value of the z values corresponding to all points, the variance of the z values corresponding to all points, etc., the application does not limit this, as long as each The eigenvalue of the grid can represent the undulation degree of the grid corresponding to the road surface.
  • the point, line, and surface features of the road surface can be determined respectively.
  • the point feature can correspond to one or a group of grids, and a group of grids can include several adjacent grids.
  • the point feature can correspond to a square area composed of four adjacent grids, or can correspond to a square area composed of four adjacent grids. For an area composed of three adjacent grids, this application does not limit the selection method and quantity of several adjacent grids.
  • the value of a point feature can be the z-value difference between the highest point and the lowest point of all point cloud data in one or a set of grids, and when the value of the point feature exceeds a predetermined threshold (for example, 10cm), a grid can be determined
  • a predetermined threshold for example, 10cm
  • the road surface corresponding to a group of grids is uneven, for example, there is a raised manhole cover, a certain pothole, a speed bump, etc. on the road surface.
  • Virtual obstacles can be set at one or a group of grids corresponding to point features to assist the determination of subsequent strategies.
  • the range of a grid or a group of grids whose point feature value exceeds a predetermined threshold may be set as the range where the virtual obstacle is located.
  • the line feature can correspond to one or more columns of continuous grids along the direction of the lane, and can represent the undulation of the road surface on a certain lane.
  • the value of the line feature can be the variance of the z values corresponding to all points in one or more consecutive grids along the direction of the lane, and when the value of the line feature exceeds a predetermined threshold (for example, 10cm 2 ), it can be determined
  • a predetermined threshold for example, 10cm 2
  • the surface feature can correspond to all grids in the region of interest.
  • the value of the surface feature can be the variance of z values corresponding to all points in the region of interest, in all grids, and when the value of the surface feature exceeds a predetermined threshold ( For example, when the vehicle is 10 cm 2 ), it can be determined that the road surface undulation of all lanes in front of the vehicle is large.
  • the values of point features, line features, and surface features can also be other values besides the above.
  • the z-value difference between the highest point and the lowest point, the average value of the z-values corresponding to all points, the variance of the z-values corresponding to all points, etc. for example, the value of the point feature can also be The variance of the z-values of all point cloud data in a grid or set of grids.
  • the value of point, line and surface features can be directly calculated according to the point cloud data in the corresponding grid, or can be calculated according to the eigenvalues of the corresponding grid.
  • the surface feature value can be determined according to the average value of the z values corresponding to each grid.
  • the point cloud data corresponds to the road in front of the vehicle
  • the vertical acceleration is obtained after the vehicle passes the current road, that is, for a certain section of road
  • the corresponding vehicle vertical acceleration data Compared with the point cloud data, it is obtained with a lag. Therefore, when determining the point, line, and surface features of the road surface, the point, line, and surface features of the corresponding road surface can be obtained according to the obtained point cloud data, and then obtained according to the obtained vertical acceleration pair. The point, line, and surface features of the corresponding road surface are corrected.
  • the values of the obtained point features, line features, and surface features can be corrected according to the vertical acceleration of the vehicle.
  • the vertical acceleration of the vehicle changes sharply
  • the rate of change of the vertical acceleration exceeds a predetermined threshold, such as exceeding 1m/s 3
  • the value of the corresponding point feature, line feature or surface feature can be modified, so that the modified
  • the value can more truly describe the degree of undulation of the road surface.
  • the greater the value of the point feature, line feature or surface feature the greater the degree of undulation of the road surface.
  • the rate of change of the vertical acceleration exceeds a predetermined threshold
  • the value of the corresponding point feature, line feature, or surface feature is proportionally enlarged.
  • step S303 the automatic driving system generates relevant strategies from three angles: lateral, longitudinal and vertical according to the determined features and vertical acceleration data.
  • the horizontal direction corresponds to the direction (Y direction) perpendicular to the driving direction on the road surface
  • the longitudinal direction corresponds to the driving direction (X direction)
  • the vertical direction corresponds to the direction perpendicular to the road surface (Z direction).
  • the automatic driving can be performed without violating traffic rules and avoiding obstacles on the road.
  • the system can generate relevant automatic driving strategies from the three dimensions of horizontal, vertical and vertical to deal with possible uneven road conditions, thereby improving the safety and comfort of the automatic driving process and reducing the structure of the vehicle caused by vibration. Fatigue damage.
  • Fig. 5 shows a schematic diagram of determining an automatic driving strategy according to an embodiment of the present application. As shown in Figure 5, in the horizontal dimension, the strategy of changing lanes and avoiding traffic within lanes can be used to deal with uneven road conditions.
  • the line features of the point cloud data corresponding to different lanes it can be determined to drive in the lane with the flattest road surface (that is, change lanes).
  • the lane with the smallest corresponding eigenvalue can be selected for driving.
  • the driving lane After determining the driving lane, it can be determined whether there is a virtual obstacle on the lane according to the point features on the lane (such as a virtual obstacle whose value of the above-mentioned corresponding point feature exceeds a grid or a group of grid ranges of a predetermined threshold ), in the presence of a virtual obstacle, make the vehicle bypass the virtual obstacle in the lane (that is, avoid the road), for example, an optimization method can be used to plan the path in the road without other real obstacles (such as other vehicles, pedestrians, cones, etc.), the distance between the path and the virtual obstacle is used as a constraint, so that the generated path is as far away from the virtual obstacle as possible, and in the case of other real obstacles, priority avoidance real obstacles.
  • This application can also use other methods to avoid traffic in the road.
  • speed limit can represent the current speed limit of the vehicle
  • speed max and speed min can be preset, respectively representing the upper limit and lower limit of the vehicle's driving speed range when the road surface is uneven
  • T 1 , T 2 , T 3 can respectively represent the normalized point eigenvalues, line eigenvalues and surface eigenvalues
  • the value range of T 1 , T 2 , T 3 is 0-1, and the maximum value of T 1 , T 2 , T 3
  • the obtained speed limit is (1- ⁇ )*speed max , and its value becomes smaller as the degree of road undulation increases (that is, the larger ⁇ is). If (1- ⁇ )*speed max is smaller than speed min , the obtained speed limit is speed min . Thereby, the vibration of the vehicle body can be further reduced in the case of an uneven road
  • this application does not limit the method of determining the current speed limit of the vehicle, as long as the greater the undulation of the road surface, the lower the corresponding upper limit of the current speed limit of the vehicle, for example, in formula (1) , it is also possible to use only one of T 1 , T 2 , T 3 or use any two of T 1 , T 2 , T 3 to determine the value of ⁇ .
  • the inertial measurement unit IMU it can be determined whether the vehicle is on an uneven road.
  • the parameters such as suspension kinematics K (kinematic) characteristics and suspension elastic kinematics C (compliance) characteristics are adjusted to alleviate the vertical vibration of the vehicle on uneven road conditions.
  • the K characteristic and C of the active suspension can be adjusted by adjusting the stiffness and damping of the active suspension. characteristics, thereby restraining the movement of the vehicle body and maintaining the stability of the vehicle body, the application does not limit the method of adjusting the stiffness and damping of the dynamic suspension.
  • the automatic driving system can simultaneously determine whether there is an uneven road surface in the horizontal, vertical and vertical dimensions, and execute related strategies in the above dimensions, or only determine and execute one or two of them; , There can also be a priority relationship between the vertical and vertical dimensions. For example, the priority in the horizontal dimension is the highest, and the priority in the vertical dimension is the lowest. The automatic driving system can determine whether there is a road in the corresponding dimension according to the priority between the dimensions. Uneven situations and execute dimensionally related strategies.
  • Step S304 the automatic driving system controls the vehicle to perform automatic driving according to the generated policy, and displays it on the human-machine interface HMI.
  • the automatic driving system can plan the driving path of the vehicle during automatic driving.
  • the automatic driving system can constrain the speed of the vehicle during automatic driving.
  • the automatic driving system can adjust the K characteristics and C characteristics of the active suspension of the vehicle during automatic driving.
  • the automatic driving system can realize the automatic driving of the vehicle on uneven road conditions by sending control signals to the vehicle.
  • the vehicle when it adopts relevant strategies due to uneven road surfaces, it can be displayed on the human-machine interface HMI to remind the drivers and passengers, so that the drivers and passengers can keep abreast of the current road conditions and the automatic driving strategies adopted, so as to improve driving efficiency. Crew trust.
  • Fig. 6 shows a schematic diagram of visual icon prompting on a human-machine interface HMI according to an embodiment of the present application. As shown in Figure 6, when the vehicle adopts a relevant strategy due to uneven road surface, the icon of the uneven road surface can change from gray to highlighted.
  • a text prompt of "uneven road surface” may also be directly displayed on the screen.
  • Table 1 shows an example of voice broadcast on the human-machine interface HMI according to an embodiment of the present application.
  • the HMI gives priority to voice broadcasting.
  • the content is "the road in the front lane is uneven, change lanes and detour".
  • the broadcast after the broadcast is finished, it can broadcast "the road in the front lane is uneven, detour in the lane”.
  • the above voice broadcast process is only an example, and other methods or other broadcast contents may also be adopted.
  • prompts can be made by playing a video.
  • Fig. 7 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • the method can be used in the above-mentioned vehicle, for example, it can be executed by an automatic driving system in the vehicle. As shown in Figure 7, the method includes:
  • Step S701 acquiring data collected by sensors, the data including point cloud data in front of the vehicle and acceleration data of the vehicle in a direction perpendicular to the road surface;
  • Step S702 determining a feature value according to the point cloud data, and the feature value is used to indicate the undulation degree of the road surface in front of the vehicle;
  • Step S703 at least according to the acceleration data and/or the characteristic value, determine the automatic driving strategy of the vehicle.
  • the current unevenness of the road surface can be accurately identified.
  • the above eigenvalues are used to determine the automatic driving strategy of the vehicle, which can reduce the impact of unexpected steering caused by uneven roads on the precise control of the vehicle, improve the safety and comfort of the vehicle during automatic driving, and at the same time reduce the , Vehicle sensors and other structural damage due to vibration, thereby improving its reliability and life durability.
  • the data collected by the sensor may be obtained by the vehicle from the sensor, or may be sent to the server by the sensor and obtained from the server.
  • the acceleration data of the vehicle in the direction perpendicular to the road surface may be the above-mentioned vertical acceleration data, and the larger the characteristic value, the greater the undulation of the road surface in front of the vehicle.
  • step S701 refers to step S301 in FIG. 3, for an example of step S702, refer to step S302 in FIG. 3, and for an example of step S703, refer to step S303 in FIG.
  • the feature value is determined according to a coordinate value of the point cloud data in a direction perpendicular to the road surface.
  • the coordinate value of the point cloud data in the direction perpendicular to the road surface can reflect the degree of undulation of the road surface. According to the embodiment of the present application, the characteristic value can better measure the unevenness of the road surface.
  • the coordinate value in the direction perpendicular to the road surface may be, for example, the value in the Z direction mentioned above.
  • the feature value is based on the z value of the highest point in the processed point cloud data, the z value of the lowest point, the difference between the z value of the highest point and the lowest point, the average value of z values, z
  • One or more of the variance of the z value represents the coordinate value of the point cloud data in the direction perpendicular to the road surface.
  • the embodiment of the present application it is possible to flexibly select the method of determining the characteristic value according to the needs, so that the degree of road surface undulation indicated by the characteristic value can be more in line with the actual situation of the vehicle during automatic driving.
  • determining the feature value according to the point cloud data includes: determining at least one of the first feature value, the second feature value and the third feature value according to the point cloud data, the The first eigenvalue is used to indicate the undulation degree of the road surface in front of the vehicle and the local area in each lane, and the second eigenvalue is used to indicate the undulation degree of the road surface in front of the vehicle and the road surface of each lane.
  • the three eigenvalues are used to indicate the degree of undulation of the road surface in front of the vehicle and the road surface of all lanes.
  • the first eigenvalue, the second eigenvalue, and the third eigenvalue by determining at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue, and the first eigenvalue, the second eigenvalue, and the third eigenvalue are obtained from points, lines, and surfaces respectively
  • the three angles indicate the degree of undulation of the road surface, which can more accurately and comprehensively reflect the unevenness of the road surface, thus providing a targeted basis for the subsequent determination of automatic driving strategies.
  • the first eigenvalue may be the value of the above-mentioned point feature
  • the second eigenvalue may be the value of the above-mentioned line feature
  • the third eigenvalue may be the value of the above-mentioned surface feature.
  • the local area may include an area of any size in the lane. The application does not limit the size of the local area, each lane, and the entire road surface. For determining each lane road surface corresponding to the first eigenvalue and each local area corresponding to the second eigenvalue And the manner of the range of all road surfaces corresponding to the third feature value is not limited.
  • Fig. 8 shows a flowchart of an automatic driving method according to an embodiment of the present application. As shown in FIG. 8, at least according to the acceleration data and/or the characteristic value, the automatic driving strategy of the vehicle is determined, including:
  • Step S801 according to the second eigenvalue, determine the target lane corresponding to the second smallest eigenvalue
  • Step S802 determining a strategy for the vehicle to switch to the target lane.
  • the strategy for the vehicle to switch to the target lane is determined, so that the vehicle can avoid the lane with a large degree of undulation to avoid uneven road conditions, Thereby improving the safety and comfort of the vehicle and reducing related structural damage.
  • the second eigenvalue is the smallest, which may indicate that the fluctuation degree of the corresponding lane is the smallest, and the target lane may be a lane in the above-mentioned region of interest.
  • steps S801-S802 reference may be made to the description about generating an automatic driving strategy from the lateral dimension in step S303 in FIG. 3 above.
  • Fig. 9 shows a flowchart of an automatic driving method according to an embodiment of the present application. As shown in FIG. 9, at least according to the acceleration data and/or the characteristic value, the automatic driving strategy of the vehicle is determined, including:
  • Step S901 according to the first feature value, determine the target local area corresponding to the first feature value exceeding a threshold
  • Step S902 determining a strategy for the vehicle to avoid the target local area.
  • the vehicle avoidance target local strategy by determining the target local area corresponding to the first eigenvalue exceeding the threshold according to the first eigenvalue, the basis for determining the vehicle avoidance target local strategy, so that the vehicle can avoid local unevenness such as potholes in the driveway road surface to avoid uneven road conditions, thereby improving vehicle safety and comfort and reducing related structural damage.
  • the threshold value can be preset according to needs, and the first characteristic value exceeds the threshold value, which can indicate that the fluctuation degree of the target local area is greater than the preset range, and the target local area can, for example, correspond to the area of the virtual obstacle mentioned above.
  • the way of avoiding the local area of the target is not limited, for example, the way of optimizing with the distance between the planned path and the virtual obstacle as a constraint can be referred to above.
  • steps S901-S902 reference may be made to the description about generating an automatic driving strategy from the lateral dimension in step S303 in FIG. 3 above.
  • determining the automatic driving strategy of the vehicle includes: according to the first characteristic value, the second characteristic value, and the third characteristic value At least one of, determine the strategy of limiting the vehicle speed, wherein, the upper limit value of the vehicle's limited driving speed, and at least one of the first characteristic value, the second characteristic value, and the third characteristic value The maximum value is negatively correlated.
  • the vehicle by determining a strategy for limiting the vehicle speed according to at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue, the vehicle can be decelerated before encountering an uneven road,
  • the upper limit value of the limited driving speed of the vehicle negatively correlated with the maximum value of at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue when determining the strategy for limiting the vehicle speed. It can be realized that the upper limit of the speed limit of the vehicle is smaller when the degree of road surface undulation is greater, so that the vibration amplitude of the vehicle can be further reduced, the safety and comfort of the vehicle can be improved, and related structural damage can be reduced.
  • the present application does not limit the manner of determining the upper limit value of the limited driving speed of the vehicle, for example, the above formula (1) may be referred to.
  • generating a strategy for instructing the vehicle to perform automatic driving according to the acceleration data and/or the feature value includes: when the feature value exceeds a threshold, or the acceleration data If the rate of change of exceeds a threshold, determine a strategy for instructing to adjust the parameters of the active suspension of the vehicle.
  • the vehicle by determining a strategy for instructing to adjust the parameters of the active suspension of the vehicle when the characteristic value exceeds the threshold value or the rate of change of the acceleration data exceeds the threshold value, it can be realized that the vehicle passes through the uneven road. , adjust the parameters of the active suspension in advance, so as to reduce the vibration amplitude of the vehicle when passing the uneven road, improve the safety and comfort of the vehicle, and reduce related structural damage.
  • the characteristic value exceeds the threshold value, which may be that at least one of the first characteristic value, the second characteristic value, and the third characteristic value exceeds the corresponding threshold value.
  • the present application does not limit the manner of adjusting the parameters of the active suspension of the vehicle.
  • determining the feature value according to the point cloud data includes: when the rate of change of the acceleration data exceeds a predetermined threshold, adjusting the determined feature value to obtain the adjusted Eigenvalues, the adjusted eigenvalues are greater than the unadjusted eigenvalues.
  • the adjusted eigenvalues can reflect the current unevenness of the road more truly, so that the adjusted eigenvalues can be more accurate.
  • the eigenvalues generate a more accurate automatic driving strategy, thereby further improving the safety and comfort of the vehicle during automatic driving, and reducing related structural damage.
  • Adjusting the determined feature value may be adjusting at least one of the first feature value, the second feature value, and the third feature value.
  • Fig. 10 shows a flowchart of an automatic driving method according to an embodiment of the present application.
  • determine feature value according to described point cloud data comprise:
  • Step S1001 dividing the region of interest in the road ahead of the vehicle into grids
  • Step S1002 deleting point cloud data corresponding to obstacles in the point cloud data in the region of interest, to obtain processed point cloud data;
  • Step S1003 determining the feature value according to the processed point cloud data in at least one grid.
  • the eigenvalues can reflect the unevenness of the road surface more accurately, so that better Auxiliary generation of relevant automatic driving policies.
  • the application does not limit the size of the divided grid, for example, it can be determined according to the density of point cloud data, and obstacles can be stationary or moving objects in the area of interest, such as other vehicles on the road, pedestrians, cones, Traffic lights and more.
  • step S1001-step S1003 For an example of step S1001-step S1003, reference may be made to the relevant description in step S302 in FIG. 3 above.
  • determining the feature value according to the point cloud data includes: determining the characteristic value according to the coordinate value of the point cloud data in one or a group of grids in any lane in the direction perpendicular to the road surface. According to the eigenvalues of the local area corresponding to the one or a set of grid data, the first eigenvalues are obtained; and/or according to the point cloud data in one or more columns of grids in any lane direction in a direction perpendicular to the road surface Determine the eigenvalue corresponding to any one of the lanes to obtain the second eigenvalue; and/or determine the second eigenvalue according to the coordinate values of the point cloud data in all grids in the direction perpendicular to the road surface Three eigenvalues.
  • the angle of the surface measures the unevenness of the road surface.
  • the method further includes: according to the strategy, generating prompt information for the undulation degree of the road ahead of the vehicle and/or the automatic driving strategy of the vehicle, the prompt information includes text, At least one of image, audio, and video.
  • the driver and passengers by generating the prompt information, the driver and passengers can know the current road conditions and the relevant strategies adopted by the vehicle in a timely manner, thereby improving the trust of the drivers and passengers.
  • the present application does not limit the method of generating the prompt information, for example, the prompt may be provided in ways other than the above-mentioned text, image, video, and audio, and the present application does not limit the content included in the prompt information.
  • step S304 in FIG. 3 For an example of the above process, refer to step S304 in FIG. 3 .
  • Fig. 11 shows a structural diagram of an automatic driving device according to an embodiment of the present application. As shown in Figure 11, the device 1100 includes:
  • An acquisition module 1101 configured to acquire data collected by sensors, the data including point cloud data in front of the vehicle and acceleration data of the vehicle in a direction perpendicular to the road surface;
  • the first determination module 1102 is configured to determine a feature value according to the point cloud data, and the feature value is used to indicate the undulation degree of the road surface in front of the vehicle;
  • the second determination module 1103 is configured to determine an automatic driving strategy of the vehicle at least according to the acceleration data and/or the characteristic value.
  • the current unevenness of the road surface can be accurately identified.
  • the above eigenvalues are used to determine the automatic driving strategy of the vehicle, which can reduce the impact of unexpected steering caused by uneven roads on the precise control of the vehicle, improve the safety and comfort of the vehicle during automatic driving, and at the same time reduce the , Vehicle sensors and other structural damage due to vibration, thereby improving its reliability and life durability.
  • the feature value is determined according to the coordinate value of the point cloud data in a direction perpendicular to the road surface.
  • the coordinate value of the point cloud data in the direction perpendicular to the road surface can reflect the degree of undulation of the road surface. According to the embodiment of the present application, the characteristic value can better measure the unevenness of the road surface.
  • the feature value is based on the z value of the highest point in the processed point cloud data, the z value of the lowest point, the difference between the z value of the highest point and the lowest point, the average value of z values, z
  • One or more of the variance of the z value represents the coordinate value of the point cloud data in the direction perpendicular to the road surface.
  • the embodiment of the present application it is possible to flexibly select the method of determining the characteristic value according to the needs, so that the degree of road surface undulation indicated by the characteristic value can be more in line with the actual situation of the vehicle during automatic driving.
  • determining the feature value according to the point cloud data includes: determining at least one of the first feature value, the second feature value and the third feature value according to the point cloud data, the The first eigenvalue is used to indicate the undulation degree of the road surface in front of the vehicle and the local area in each lane, and the second eigenvalue is used to indicate the undulation degree of the road surface in front of the vehicle and the road surface of each lane.
  • the three eigenvalues are used to indicate the degree of undulation of the road surface in front of the vehicle and the road surface of all lanes.
  • the first eigenvalue, the second eigenvalue, and the third eigenvalue by determining at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue, and the first eigenvalue, the second eigenvalue, and the third eigenvalue are obtained from points, lines, and surfaces respectively
  • the three angles indicate the degree of undulation of the road surface, which can more accurately and comprehensively reflect the unevenness of the road surface, thus providing a targeted basis for the subsequent determination of automatic driving strategies.
  • determining the automatic driving strategy of the vehicle includes: according to the second characteristic value, determining the vehicle corresponding to the second smallest characteristic value. A target lane; determining a strategy for the vehicle to switch to the target lane.
  • the strategy for the vehicle to switch to the target lane is determined, so that the vehicle can avoid the lane with a large degree of undulation to avoid uneven road conditions, Thereby improving the safety and comfort of the vehicle and reducing related structural damage.
  • determining the automatic driving strategy of the vehicle includes: according to the first feature value, determining that the corresponding first feature value exceeds a threshold a target local area; determining a strategy for the vehicle to avoid the target local area.
  • the vehicle avoidance target local strategy by determining the target local area corresponding to the first eigenvalue exceeding the threshold according to the first eigenvalue, the basis for determining the vehicle avoidance target local strategy, so that the vehicle can avoid local unevenness such as potholes in the driveway road surface to avoid uneven road conditions, thereby improving vehicle safety and comfort and reducing related structural damage.
  • determining the automatic driving strategy of the vehicle includes: according to the first characteristic value, the second characteristic value, and the third characteristic value At least one of, determine the strategy of limiting the vehicle speed, wherein, the upper limit value of the vehicle's limited driving speed, and at least one of the first characteristic value, the second characteristic value, and the third characteristic value The maximum value is negatively correlated.
  • the vehicle by determining a strategy for limiting the vehicle speed according to at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue, the vehicle can be decelerated before encountering an uneven road,
  • the upper limit value of the limited driving speed of the vehicle negatively correlated with the maximum value of at least one of the first eigenvalue, the second eigenvalue, and the third eigenvalue when determining the strategy for limiting the vehicle speed. It can be realized that the upper limit of the speed limit of the vehicle is smaller when the degree of road surface undulation is greater, so that the vibration amplitude of the vehicle can be further reduced, the safety and comfort of the vehicle can be improved, and related structural damage can be reduced.
  • determining the automatic driving strategy of the vehicle includes: when the characteristic value exceeds a threshold, or the rate of change of the acceleration data If the threshold value is exceeded, a strategy for adjusting the parameters of the active suspension of the vehicle is determined.
  • the vehicle by determining a strategy for instructing to adjust the parameters of the active suspension of the vehicle when the characteristic value exceeds the threshold value or the rate of change of the acceleration data exceeds the threshold value, it can be realized that the vehicle passes through the uneven road. , adjust the parameters of the active suspension in advance, so as to reduce the vibration amplitude of the vehicle when passing the uneven road, improve the safety and comfort of the vehicle, and reduce related structural damage.
  • determining the feature value according to the point cloud data includes: when the rate of change of the acceleration data exceeds a predetermined threshold, adjusting the determined feature value to obtain the adjusted Eigenvalues, the adjusted eigenvalues are greater than the unadjusted eigenvalues.
  • the adjusted eigenvalues can reflect the current unevenness of the road more truly, so that the adjusted eigenvalues can be more accurate.
  • the eigenvalues generate a more accurate automatic driving strategy, thereby further improving the safety and comfort of the vehicle during automatic driving, and reducing related structural damage.
  • determining the feature value according to the point cloud data includes: dividing the region of interest in the road ahead of the vehicle into grids; deleting the point cloud data in the region of interest, and obstacles The point cloud data corresponding to the object is obtained to obtain the processed point cloud data; and the feature value is determined according to the processed point cloud data in at least one grid.
  • the eigenvalues can reflect the unevenness of the road surface more accurately, so that better Auxiliary generation of relevant automatic driving policies.
  • determining the feature value according to the point cloud data includes: determining the characteristic value according to the coordinate value of the point cloud data in one or a group of grids in any lane in the direction perpendicular to the road surface. According to the eigenvalues of the local area corresponding to the one or a set of grid data, the first eigenvalues are obtained; and/or according to the point cloud data in one or more columns of grids in any lane direction in a direction perpendicular to the road surface Determine the eigenvalue corresponding to any one of the lanes to obtain the second eigenvalue; and/or determine the second eigenvalue according to the coordinate values of the point cloud data in all grids in the direction perpendicular to the road surface Three eigenvalues.
  • the angle of the surface measures the unevenness of the road surface.
  • the device further includes: a generation module, configured to generate prompt information for the undulation degree of the road surface ahead of the vehicle and/or the automatic driving strategy of the vehicle according to the strategy, the The prompt information includes at least one of text, image, audio and video.
  • the driver and passengers by generating the prompt information, the driver and passengers can know the current road conditions and the relevant strategies adopted by the vehicle in a timely manner, thereby improving the trust of the drivers and passengers.
  • Fig. 12 shows a structural diagram of an automatic driving device according to an embodiment of the present application.
  • the automatic driving device can be applied to the vehicle shown in FIG. 1 , and execute the functions in the automatic driving method shown in any one of the above-mentioned FIGS. 3-10 .
  • the automatic driving device may be the above-mentioned vehicle, or a chip (system) or other components or components that may be arranged inside the vehicle.
  • the automatic driving device may also be the above-mentioned automatic driving device 1100 .
  • an automatic driving device 1200 may include a processor 1201 and a transceiver 1202 .
  • the automatic driving device 1200 may include a memory 1203 .
  • the processor 1201 is coupled with the transceiver 1202 and the memory 1203, such as may be connected through a communication bus.
  • the components of the automatic driving device 1200 will be specifically introduced below with reference to FIG. 12 .
  • the above-mentioned processor 1201 is the control center of the automatic driving device 1200, and may be one processor, or a general term for multiple processing elements.
  • the processor 1201 is one or more central processing units (central processing unit, CPU), may also be a specific integrated circuit (application specific integrated circuit, ASIC), or is configured to implement one or more An integrated circuit, for example: one or more microprocessors (digital signal processor, DSP), or, one or more field programmable gate arrays (field programmable gate array, FPGA).
  • the processor 1201 can execute various functions of the automatic driving device 1200 by running or executing software programs stored in the memory 1203 and calling data stored in the memory 1203 .
  • the processor 1201 may include one or more CPUs, such as CPU0 and CPU1 shown in FIG. 12 .
  • the automatic driving device 1200 may also include multiple processors, such as the processor 1201 and the processor 1204 shown in FIG. 12 .
  • processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU).
  • a processor herein may refer to one or more communication devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • the above-mentioned transceiver 1202 is used for communication with other devices.
  • the automatic driving device 1200 is a vehicle, and the transceiver 1202 can be used to communicate with the server.
  • the transceiver 1202 may include a receiver and a transmitter (not separately shown in FIG. 12 ). Wherein, the receiver is used to realize the receiving function, and the transmitter is used to realize the sending function.
  • the transceiver 1202 may be integrated with the processor 1201 or exist independently, and be coupled with the processor 1201 through an input/output port (not shown in FIG. 12 ) of the automatic driving device 1200 .
  • the above-mentioned memory 1203 can be used to store the software program for implementing the solution of the present application, and the execution is controlled by the processor 1201.
  • the specific implementation please refer to the above-mentioned method embodiment, which will not be repeated here.
  • the memory 1203 may be a read-only memory (read-only memory, ROM) or other types of static storage communication devices that can store static information and instructions, or a random access memory (random access memory, RAM) that can store information and instructions
  • ROM read-only memory
  • RAM random access memory
  • Other types of dynamic storage communication devices can also be electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical discs Storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage communication devices, or can be used to carry or store desired programs in the form of instructions or data structures code and any other medium that can be accessed by a computer, without limitation.
  • the memory 1203 can be integrated with the processor 1201 or exist independently, and is coupled with the processor 1201 through an input/output port (not shown in FIG. 12 ) of the automatic driving device 1200 .
  • the structure of the automatic driving device 1200 shown in FIG. 12 does not constitute a limitation on the implementation of the automatic driving device.
  • the actual automatic driving device may include more or fewer components than shown in the figure, or Combining certain parts, or different arrangements of parts.
  • An embodiment of the present application provides an automatic driving device, including: a processor and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above method when executing the instructions.
  • Embodiments of the present application provide a terminal device, and the terminal device can execute the foregoing method.
  • An embodiment of the present application provides a non-volatile computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the foregoing method is implemented.
  • An embodiment of the present application provides a computer program product, including computer-readable codes, or a non-volatile computer-readable storage medium bearing computer-readable codes, when the computer-readable codes are stored in a processor of an electronic device When running in the electronic device, the processor in the electronic device executes the above method.
  • a computer readable storage medium may be a tangible device that can retain and store instructions for use by an instruction execution device.
  • a computer readable storage medium may be, for example, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Non-exhaustive list of computer-readable storage media include: portable computer disk, hard disk, random access memory (Random Access Memory, RAM), read only memory (Read Only Memory, ROM), erasable Electrically Programmable Read-Only-Memory (EPROM or flash memory), Static Random-Access Memory (Static Random-Access Memory, SRAM), Portable Compression Disk Read-Only Memory (Compact Disc Read-Only Memory, CD -ROM), Digital Video Disc (DVD), memory sticks, floppy disks, mechanically encoded devices such as punched cards or raised structures in grooves with instructions stored thereon, and any suitable combination of the foregoing .
  • RAM Random Access Memory
  • ROM read only memory
  • EPROM or flash memory erasable Electrically Programmable Read-Only-Memory
  • Static Random-Access Memory SRAM
  • Portable Compression Disk Read-Only Memory Compact Disc Read-Only Memory
  • CD -ROM Compact Disc Read-Only Memory
  • DVD Digital Video Disc
  • Computer readable program instructions or codes described herein may be downloaded from a computer readable storage medium to a respective computing/processing device, or downloaded to an external computer or external storage device over a network, such as the Internet, local area network, wide area network, and/or wireless network.
  • the network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • a network adapter card or a network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .
  • Computer program instructions for performing the operations of the present application may be assembly instructions, instruction set architecture (Instruction Set Architecture, ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or in one or more source or object code written in any combination of programming languages, including object-oriented programming languages—such as Smalltalk, C++, etc., and conventional procedural programming languages—such as the “C” language or similar programming languages.
  • Computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement.
  • the remote computer can be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it can be connected to an external computer such as use an Internet service provider to connect via the Internet).
  • electronic circuits such as programmable logic circuits, field-programmable gate arrays (Field-Programmable Gate Array, FPGA) or programmable logic arrays (Programmable Logic Array, PLA), the electronic circuit can execute computer-readable program instructions, thereby realizing various aspects of the present application.
  • These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine such that when executed by the processor of the computer or other programmable data processing apparatus , producing an apparatus for realizing the functions/actions specified in one or more blocks in the flowchart and/or block diagram.
  • These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause computers, programmable data processing devices and/or other devices to work in a specific way, so that the computer-readable medium storing instructions includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks in flowcharts and/or block diagrams.
  • each block in a flowchart or block diagram may represent a module, a portion of a program segment, or an instruction that includes one or more Executable instructions.
  • the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block in the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented with hardware (such as circuits or ASIC (Application Specific Integrated Circuit, application-specific integrated circuit)), or can be implemented with a combination of hardware and software, such as firmware.
  • hardware such as circuits or ASIC (Application Specific Integrated Circuit, application-specific integrated circuit)
  • firmware such as firmware

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种自动驾驶方法、装置和存储介质，该方法包括：获取传感器采集的数据，数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据(S701)；根据点云数据确定特征值，特征值用于指示车辆前方路面的起伏程度(S702)；至少根据加速度数据和/或所述特征值，确定车辆的自动驾驶策略(s703)。可以准确地识别出当前路面不平的情况，降低路面不平产生的非预期转向等对车辆精准控制的影响，提高车辆在自动驾驶过程中的安全性和舒适性，同时，可以减小车辆、车辆上的传感器等因振动产生的结构性损伤，从而提升其可靠性和寿命耐久性。

Description

自动驾驶方法、装置和存储介质 技术领域
本申请涉及自动驾驶技术领域，尤其涉及一种自动驾驶方法、装置和存储介质。
背景技术
随着自动驾驶技术的发展，车辆不仅需要在平坦路面上实现自动驾驶，也需要应对坑洼路面、破损路面、铺设减速带的路面等路面不平的情况。
目前，相关技术侧重于应对行人、红绿灯等路面上的障碍物，而对于路面不平的情况，通常是在车辆振动后，依靠车辆自身的悬架***滞后缓解，导致车辆的安全性和舒适性低下。
发明内容
有鉴于此，提出了一种自动驾驶方法、装置和存储介质。
第一方面，本申请的实施例提供了一种自动驾驶方法，该方法包括：获取传感器采集的数据，所述数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据；根据所述点云数据确定特征值，所述特征值用于指示车辆前方路面的起伏程度；至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略。
根据本申请实施例，通过获取传感器采集的车辆前方数据，根据点云数据确定特征值，可以准确地识别出当前路面不平的情况，通过根据车辆在垂直于路面方向上的加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，可以降低路面不平产生的非预期转向等对车辆精准控制的影响，提高车辆在自动驾驶过程中的安全性和舒适性，同时，可以减小车辆、车辆上的传感器等因振动产生的结构性损伤，从而提升其可靠性和寿命耐久性。
根据第一方面，在所述自动驾驶方法的第一种可能的实现方式中，特征值根据所述点云数据在垂直于路面方向上的坐标值确定。
点云数据在垂直于路面方向上的坐标值可以反映路面的起伏程度，根据本申请实施例，可以使得特征值可以更好的衡量路面不平的情况。
根据第一方面，在所述自动驾驶方法的第二种可能的实现方式中，根据所述点云数据确定特征值，包括：根据所述点云数据，确定第一特征值、第二特征值和第三特征值中的至少一种，所述第一特征值用于指示所述车辆前方路面中、各车道内局部区域的起伏程度，所述第二特征值用于指示所述车辆前方路面中、各车道路面的起伏程度，所述第三特征值用于指示所述车辆前方路面中、全部车道路面的起伏程度。
根据本申请实施例，通过确定第一特征值、第二特征值和第三特征值中的至少一种，且第一特征值、第二特征值和第三特征值分别从点、线、面三个角度指示路面的起伏程度，可以更精准全面的体现路面不平的情况，从而为后续确定自动驾驶策略提供了针对性的依据。
根据第一方面的第二种可能的实现方式，在所述自动驾驶方法的第三种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据所述第二特征值，确定对应第二特征值最小的目标车道；确定所述车辆切换至所述目标车道的策略。
根据本申请实施例，通过根据第二特征值，确定对应第二特征值最小的目标车道，确定车辆切换至目标车道的策略，使得车辆可以规避起伏程度大的车道，以避免路面不平的情况，从而提高车辆的安全性和舒适性，减小相关结构性损伤。
根据第一方面的第二种或第三种可能的实现方式，在所述自动驾驶方法的第四种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据所述第一特征值，确定对应第一特征值超过阈值的目标局部区域；确定所述车辆避行所述目标局部区域的策略。
根据本申请实施例，通过根据第一特征值，确定对应第一特征值超过阈值的目标局部区域，确定车辆避行目标局部策略的依据，使得车辆可以绕开车道内的坑洼等局部不平的路面，以避免路面不平的情况，从而提高车辆的安全性和舒适性，减小相关结构性损伤。
根据第一方面的第二种或第三种或第四种可能的实现方式，在所述自动驾驶方法的第五种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，其中，车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关。
根据本申请实施例，通过根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，可以实现车辆在遇到不平路面前提前减速，通过在确定限制车辆行驶速度的策略时，使得车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关，可以实现在路面起伏程度越大的情况下，车辆的限速的上限值越小，从而可以进一步减小车辆的振动幅度，提高车辆的安全性和舒适性，减小相关结构性损伤。
根据第一方面或第一方面的第一种或第二种或第三种或第四种或第五种可能的实现方式，在所述自动驾驶方法的第六种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：在所述特征值超过阈值，或所述加速度数据的变化率超过阈值的情况下，确定对所述车辆的主动悬架的参数进行调整的策略。
根据本申请实施例，通过在特征值超过阈值，或加速度数据的变化率超过阈值的情况下，确定用于指示对车辆的主动悬架的参数进行调整的策略，可以实现在车辆经过不平路面前，提前对主动悬架的参数进行调整，从而可以减小车辆在经过不平路面时的振动幅度，提高车辆的安全性和舒适性，减小相关结构性损伤。
根据第一方面，在所述自动驾驶方法的第七种可能的实现方式中，根据所述点云数据确定特征值，包括：在所述加速度数据的变化率超过预定阈值的情况下，对确定的所述特征值进行调整，得到调整后的特征值，所述调整后的特征值大于调整前的特征值。
根据本申请实施例，通过在加速度数据的变化率超过阈值的情况下，对确定的特征值进行调整，可以使得调整后的特征值更真实的反映当前路面不平的情况，使得可以更具调整后的特征值生成更加精准的自动驾驶的策略，从而进一步提高车辆在自动驾驶过程中的安全性和舒适性，减小相关结构性损伤。
根据第一方面或第一方面的第一种或第二种或第三种或第四种或第五种或第六种或第七种可能的实现方式，在所述自动驾驶方法的第八种可能的实现方式中，根据所述点云数据确定特征值，包括：将车辆前方路面中的感兴趣区域划分为网格；删除所述感兴趣区域内的点 云数据中，与障碍物对应的点云数据，得到处理后的点云数据；根据至少一个所述网格中的处理后的点云数据，确定所述特征值。
根据本申请实施例，通过将车辆前方路面中的感兴趣区域划分为网格，删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，可以减小其他车辆、行人等障碍物对于判断路面不平度的影响，通过根据至少一个所述网格中的处理后的点云数据，确定特征值，可以实现特征值更加准确的反映路面的不平情况，从而可以更好地辅助生成相关的自动驾驶的策略。
根据第一方面的第八种可能的实现方式，在所述自动驾驶方法的第九种可能的实现方式中，根据所述点云数据确定特征值，包括：根据任一车道内的一个或一组网格中的点云数据在垂直于路面方向上的坐标值，确定所述一个或一组网格数据对应的局部区域的特征值，得到所述第一特征值；和/或根据任一车道方向上的一列或多列网格中的点云数据在垂直于路面方向上的坐标值，确定所述任一车道对应的特征值，得到所述第二特征值；和/或根据所有网格中的点云数据在垂直于路面方向上的坐标值，确定所述第三特征值。
根据本申请实施例，通过根据相关网格中的点云数据在垂直于路面方向上的坐标值，确定第一特征值、第二特征值、第三特征值，可以更加精准地从点、线、面的角度衡量路面不平的情况。
根据第一方面或第一方面的第一种或第二种或第三种或第四种或第五种或第六种或第七种或第八种或第九种可能的实现方式，在所述自动驾驶方法的第十种可能的实现方式中，所述特征值根据处理后的点云数据中最高点的z值、最低点的z值、最高点和最低点的z值差、z值的平均值、z值的方差中的一种或多种确定，所述z值表示点云数据在垂直于路面方向上的坐标值。
根据本申请实施例，可以实现根据需要灵活选择确定特征值的方式，使得特征值所指示的路面起伏程度可以更加符合车辆在自动驾驶时的实际情况。
根据第一方面或第一方面的第一种或第二种或第三种或第四种或第五种或第六种或第七种或第八种或第九种或第十种可能的实现方式，在所述自动驾驶方法的第十一种可能的实现方式中，该方法还包括：根据所述策略，生成对所述车辆前方路面的起伏程度和/或所述车辆的自动驾驶策略的提示信息，所述提示信息包括文字、图像、音频、视频中的至少一种。
根据本申请实施例，通过生成提示信息，可以使得驾乘人员能过及时了解当前路面的情况、以及车辆采取的相关策略，从而可以提高驾乘人员的信任感。
第二方面，本申请的实施例提供了一种自动驾驶装置，该装置包括：获取模块，用于获取传感器采集的数据，所述数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据；第一确定模块，用于根据所述点云数据确定特征值，所述特征值用于指示车辆前方路面的起伏程度；第二确定模块，用于至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略。
根据第二方面，在所述自动驾驶装置的第一种可能的实现方式中，特征值根据所述点云数据在垂直于路面方向上的坐标值确定。
根据第二方面，在所述自动驾驶装置的第二种可能的实现方式中，根据所述点云数据确定特征值，包括：根据所述点云数据，确定第一特征值、第二特征值和第三特征值中的至少一种，所述第一特征值用于指示所述车辆前方路面中、各车道内局部区域的起伏程度，所述 第二特征值用于指示所述车辆前方路面中、各车道路面的起伏程度，所述第三特征值用于指示所述车辆前方路面中、全部车道路面的起伏程度。
根据第二方面的第二种可能的实现方式，在所述自动驾驶装置的第三种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据所述第二特征值，确定对应第二特征值最小的目标车道；确定所述车辆切换至所述目标车道的策略。
根据第二方面的第二种或第三种可能的实现方式，在所述自动驾驶装置的第四种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据所述第一特征值，确定对应第一特征值超过阈值的目标局部区域；确定所述车辆避行所述目标局部区域的策略。
根据第二方面的第二种或第三种或第四种可能的实现方式，在所述自动驾驶装置的第五种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，其中，车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关。
根据第二方面或第二方面的第一种或第二种或第三种或第四种或第五种可能的实现方式，在所述自动驾驶装置的第六种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：在所述特征值超过阈值，或所述加速度数据的变化率超过阈值的情况下，确定对所述车辆的主动悬架的参数进行调整的策略。
根据第二方面，在所述自动驾驶装置的第七种可能的实现方式中，根据所述点云数据确定特征值，包括：在所述加速度数据的变化率超过预定阈值的情况下，对确定的所述特征值进行调整，得到调整后的特征值，所述调整后的特征值大于调整前的特征值。
根据第二方面或第二方面的第一种或第二种或第三种或第四种或第五种或第六种或第七种可能的实现方式，在所述自动驾驶装置的第八种可能的实现方式中，根据所述点云数据确定特征值，包括：将车辆前方路面中的感兴趣区域划分为网格；删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，得到处理后的点云数据；根据至少一个所述网格中的处理后的点云数据，确定所述特征值。
根据第二方面的第八种可能的实现方式，在所述自动驾驶装置的第九种可能的实现方式中，根据所述点云数据确定特征值，包括：根据任一车道内的一个或一组网格中的点云数据在垂直于路面方向上的坐标值，确定所述一个或一组网格数据对应的局部区域的特征值，得到所述第一特征值；和/或根据任一车道方向上的一列或多列网格中的点云数据在垂直于路面方向上的坐标值，确定所述任一车道对应的特征值，得到所述第二特征值；和/或根据所有网格中的点云数据在垂直于路面方向上的坐标值，确定所述第三特征值。
根据第二方面或第二方面的第一种或第二种或第三种或第四种或第五种或第六种或第七种或第八种或第九种可能的实现方式，在所述自动驾驶装置的第十种可能的实现方式中，所述特征值根据处理后的点云数据中最高点的z值、最低点的z值、最高点和最低点的z值差、z值的平均值、z值的方差中的一种或多种确定，所述z值表示点云数据在垂直于路面方向上的坐标值。
根据第二方面或第二方面的第一种或第二种或第三种或第四种或第五种或第六种或第七 种或第八种或第九种或第十种可能的实现方式，在所述自动驾驶装置的第十一种可能的实现方式中，该装置还包括：生成模块，用于根据所述策略，生成对所述车辆前方路面的起伏程度和/或所述车辆的自动驾驶策略的提示信息，所述提示信息包括文字、图像、音频、视频中的至少一种。
第三方面，本申请的实施例提供了一种模自动驾驶装置，该装置包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为执行所述指令时实现上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的自动驾驶方法。
第四方面，本申请的实施例提供了一种非易失性计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的自动驾驶方法。
第五方面，本申请的实施例提供了一种终端设备，该终端设备可以执行上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的自动驾驶方法。
第六方面，本申请的实施例提供了一种计算机程序产品，包括计算机可读代码，或者承载有计算机可读代码的非易失性计算机可读存储介质，当所述计算机可读代码在电子设备中运行时，所述电子设备中的处理器执行上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的自动驾驶方法。
本申请的这些和其他方面在以下(多个)实施例的描述中会更加简明易懂。
附图说明
包含在说明书中并且构成说明书的一部分的附图与说明书一起示出了本申请的示例性实施例、特征和方面，并且用于解释本申请的原理。
图1示出根据本申请一实施例的应用场景的示意图。
图2示出根据本申请一实施例的自动驾驶装置的架构示意图。
图3示出根据本申请一实施例的自动驾驶方法的流程图。
图4示出根据本申请一实施例的确定点、线、面特征的示意图。
图5示出根据本申请一实施例的确定自动驾驶策略的示意图。
图6示出根据本申请一实施例的在人机界面HMI上进行视觉图标提示的示意图。
图7示出根据本申请一实施例的自动驾驶方法的流程图。
图8示出根据本申请一实施例的自动驾驶方法的流程图。
图9示出根据本申请一实施例的自动驾驶方法的流程图。
图10示出根据本申请一实施例的自动驾驶方法的流程图。
图11示出根据本申请一实施例的自动驾驶装置的结构图。
图12示出根据本申请一实施例的自动驾驶装置的结构图。
具体实施方式
以下将参考附图详细说明本申请的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面，但是除非特别指出，不必按比例绘制附图。
在这里专用的词“示例性”意为“用作例子、实施例或说明性”。这里作为“示例性”所 说明的任何实施例不必解释为优于或好于其它实施例。
另外，为了更好的说明本申请，在下文的具体实施方式中给出了众多的具体细节。本领域技术人员应当理解，没有某些具体细节，本申请同样可以实施。在一些实例中，对于本领域技术人员熟知的方法、手段、元件和电路未作详细描述，以便于凸显本申请的主旨。
图1示出根据本申请一实施例的应用场景的示意图。本申请一实施例的方法可用于车辆在路面上进行自动驾驶的过程，其中，在车辆行驶时可能会遇到路面不平的情况，如图1所示，路面不平的情况可以包括有凸起的井盖的路面，铺设有减速带的路面，坑洼路面，在一种可能的实施方式中，在上述场景下，本申请实施例的自动驾驶方法可以利用传感器检测出路面不平的特征，从横向、纵向和垂向三个角度分别生成相关策略，并根据相关策略实现车辆的自动驾驶，以实现车辆通过不平路面，从而降低或避免路面不平的情况对车辆的影响，以提升车辆在自动驾驶过程中的安全性和舒适性，同时减少传感器和车辆等的结构疲劳损伤，提高其可靠性和稳定性。
需要说明的是，图1中所示的路面不平的情况仅为一个示例，路面不平的情况还可以包括凹凸起伏路面、破损路面、碎石路、接缝路面等其他情况，本申请实施例的自动驾驶方法还可以用于其他场景中。
图2示出根据本申请一实施例的自动驾驶装置的架构示意图。如图2所示，自动驾驶装置可以包括车辆，车辆可以是具有无线通信功能的车辆，其中，无线通信功能可设置于该车辆的车载终端、车载模组、车载单元、芯片(***)或其他部件或组件。本申请的实施例中的车辆可以处于自动驾驶状态，即车辆完全自主驾驶，无需驾驶员的控制或仅需驾驶员的少量控制。
车辆上可以设置有至少一个传感器，如车载雷达(如毫米波雷达、激光雷达、超声波雷达等)、雨量传感器、摄像头、车姿传感器(如陀螺仪)、惯性测量单元(inertial measurement unit，IMU)、全球导航卫星***(global navigation satellite system，GNSS)等，车辆上还可以设置有其他传感器，上述传感器可以设置在一个车辆上，也可以设置在多个车辆上，通过车辆上设置的至少一个传感器，可以采集到路面的点云数据以及车辆的垂向加速度(即车辆在垂直于路面方向上的加速度数据)等数据。
车辆上还可以设置有自动驾驶***，自动驾驶***可用于根据传感器采集的数据，生成用于应对路面不平情况的自动驾驶策略，并根据生成的策略实现车辆的自动驾驶，以实现车辆通过不平路面。
车辆上还可以设置有人机界面(human machine interface，HMI)，HMI可用于通过视觉图标、语音播报方式对当前的路面情况以及自动驾驶***对车辆采取的策略进行播报，以提醒相关驾乘人员。
在一种可能的实现方式中，本申请的自动驾驶装置还可以包括服务器，服务器可以作为车载计算单元位于上述车辆上，也可以位于云端，可以是实体设备，也可以是虚拟设备如虚拟机、容器等，具有无线通信功能，其中，无线通信功能可设置于该服务器的芯片(***)或其他部件或组件。服务器和车辆可以通过无线连接的方式进行通信，例如可以通过2G/3G/4G/5G等移动通信技术，以及Wi-Fi、蓝牙、调频(frequency modulation，FM)、数传电台、卫星通信等无线通信方式进行通信，例如在测试中，服务器可承载于车辆上并与车辆通过无线连接的方式进行通信，通过车辆和服务器之间的通信，服务器可以收集一个或多个 车辆上、或是设置在道路上或其他地方的传感器采集到的数据进行计算，并将计算结果回传给对应的车辆。
以下以图2中的架构为例，对本申请实施例的自动驾驶方法的流程进行说明。
图3示出根据本申请一实施例的自动驾驶方法的流程图。如图3所示，自动驾驶方法的流程可包括：
步骤S301，车辆获取传感器采集的数据。
其中，传感器采集的数据可以包括激光雷达采集的点云数据以及IMU采集的车辆的垂向加速度数据。点云数据可以是车辆行驶过程中、车辆前方的点云数据，车辆的垂向加速度可以指车辆在垂直于路面方向上的加速度，车辆的垂向加速度可以用于表示车辆在某个路段上的垂直振动程度。
传感器可以设置在一个或多个车辆上，可由一个车辆来获取传感器采集的数据。例如，传感器可以设置在一个车辆上，传感器还可以设置在除该一个车辆外的其他车辆上，以辅助该一个车辆进行自动驾驶，例如，其他车辆上的传感器采集的点云数据可对该一个车辆采集的点云数据进行补充。在一种可能的实现方式中，传感器还可以设置在其他地方，例如传感器可以设置在道路两侧，可用于采集点云数据以对车辆上的传感器采集的点云数据进行补充。
步骤S302，自动驾驶***根据传感器采集的数据，确定路面的点、线、面特征。
其中，路面的点、线、面特征可以分别从三个特征维度表示路面的起伏程度。
图4示出根据本申请一实施例的确定点、线、面特征的示意图。如图4所示，在车辆获取到激光雷达采集的点云数据后，自动驾驶***可以首先对获取到的点云数据进行坐标转换，以得到点云数据中的每个点在笛卡尔坐标系下、以车辆的后轴中心点为原点的三维坐标，例如(x,y,z)，x、y、z可以分别表示对应的点在x轴、y轴和z轴上的坐标值，例如，在图4中，P i、P i+1和P i+2可以分别表示笛卡尔坐标系下、点云数据中的三个点，在得到点云数据对应的三维坐标后，可以标定得到参考基准路面，参考基准路面例如可以通过对点云数据进行滤波后确定，可以表示路面的水平基准面，点云数据中的各点在z轴方向上越偏离参考基准路面，可以表示该点对应的路面起伏越大，需要说明的是，参考基准路面也可以通过其他方式确定。
接着，可以从点云数据中、筛选感兴趣区域的路面的点云数据，感兴趣区域可以包括车辆前方的行驶路面(即车辆前方的车道)，并筛除感兴趣区域路面上、静止或移动的障碍物对应的点云数据。其中，可以通过随机抽样一致性算法(random sample consensus，RANSAC)对获取到的点云数据进行拟合以确定车道，从而确定感兴趣区域的边界，还可以预先设定好感兴趣区域的范围，例如，感兴趣区域可以以行驶方向上最左侧的车道线和最右侧的车道线作为左右边界，以车头以及车头前方预设距离(例如30米)作为前后边界。感兴趣区域可以包括行驶方向上的全部车道，或者部分车道(例如除应急车道之外的车道，或者除禁行车道之外的车道)。障碍物可以包括路面上的其他车辆、行人、锥桶、红绿灯等等，例如，可以将点云数据对应的三维坐标中、z轴的值在预定阈值范围外(例如50cm以外)的点对应的区域作为需要筛除的障碍物。
自动驾驶***还可以将车辆前方感兴趣区域进行网格化处理，网格的分辨率大小可以是固定的，如感兴趣区域的大小可以为30m*10m，每一个网格的分辨率为1m*1m，则感兴趣区域可以被分为300个大小一致的网格；网格的分辨率大小还可以是变化的，如在x轴方向上， 随着路面距车辆的距离越远，对应的点云越稀疏，网格的分辨率大小可以由1m*1m逐渐变大，例如逐渐变化为2m*1m。
根据每个网格内的点云数据，可以确定每个网格对应的特征值，每个网格的特征值可以例如是网格内点云数据中最高点的z值(即z轴上的值)、最低点的z值、最高点和最低点的z值差、所有点对应的z值的平均值、所有点对应的z值的方差等等，本申请对此不作限制，只要每个网格的特征值可以表示网格对应路面的起伏程度即可。
根据感兴趣区域内的至少一个网格及网格内的点云数据，可以分别确定路面的点、线、面特征。
其中，点特征可以对应于一个或一组网格，一组网格可包括相邻的几个网格，点特征可以例如对应由相邻的四个网格组成的方形区域，也可以对应由相邻的三个网格组成的区域，本申请对于相邻的几个网格的选取方式以及数量不作限定。例如，点特征的值可以是一个或一组网格中、所有点云数据的最高点和最低点的z值差，在点特征的值超过预定阈值(例如10cm)时，可以确定一个网格或一组网格对应的路面不平，例如路面上存在凸起的井盖、某一处坑洼、减速带等等，在一种可能的情况下，在确定网格对应的路面不平的情况下，可以在点特征对应的一个或一组网格处设置虚拟障碍物，以辅助后续策略的确定。例如，可以将点特征的值超过预定阈值的一个网格或者一组网格的范围设置为虚拟障碍物所在的范围。
线特征可以对应于沿车道方向上的一列或多列的连续多个网格，可以表示某一条车道上的路面起伏程度。例如，线特征的值可以是沿车道方向上的一列或多列的连续多个网格中、所有点对应的z值的方差，在线特征的值超过预定阈值(例如10cm 2)时，可以确定对应的车道路面起伏大，例如可以是碎石路等等。
面特征可以对应于感兴趣区域内的所有网格，例如，面特征的值可以是感兴趣区域内、所有网格中、所有点对应的z值的方差，在面特征的值超过预定阈值(例如10cm 2)时，可以确定车辆前方的全部车道路面起伏大。
点特征、线特征、面特征的值还可以是除上述外的其他值，例如可以根据对应网格的所有点中、最高点的z值(即z轴上的值)、最低点的z值、最高点和最低点的z值差、所有点对应的z值的平均值、所有点对应的z值的方差等等中的一种或多种确定，例如，点特征的值也可以是是一个或一组网格中、所有点云数据的z值的方差。点、线、面特征的值可以直接根据对应网格内的点云数据计算，也可以根据对应网格的特征值计算，本申请不作限制，例如在面特征值为所有点对应的z值的平均值的情况下，面特征值可以根据每个网格对应的z值的平均值确定。
在一种可能的实现方式中，由于点云数据对应于车辆前方的路面，垂向加速度是在车辆经过当前路面后得到的，也就是说，对于某一段路面，对应的车辆垂向加速度的数据相对于点云数据滞后获得，因此，在确定路面的点、线、面特征时，可以先根据获取的点云数据得到对应路面的点、线、面特征，再根据获取的垂向加速度对得到的对应路面的点、线、面特征进行修正。
其中，在获得车辆在对应网格上的垂向加速度数据后，可以根据车辆的垂向加速度对得到的点特征、线特征、面特征的值进行修正，例如，在车辆的垂向加速度变化剧烈的情况下(变化剧烈的情况例如垂向加速度的变化率超过一预定阈值，例如超过1m/s 3)，可以修改对应的点特征、线特征或面特征的值的大小，以使得修改后的值可以更真实地描述路面的起伏 程度，例如，在可以点特征、线特征或面特征的值越大，表示路面的起伏程度越大的情况下，在垂向加速度的变化率超过一预定阈值的情况下，根据变化率超出阈值的程度，等比例地放大对应的点特征、线特征或面特征的值的大小。
步骤S303，自动驾驶***根据确定的特征以及垂向加速度数据，从横向、纵向和垂向三个角度生成相关的策略。其中横向对应路面中与行驶方向垂直的方向(Y方向)，纵向对应行驶方向(X方向)，垂向对应垂直于路面的方向(Z方向)。
根据步骤S301中获取到的垂向加速度数据，和步骤S302中确定的点特征、线特征和面特征对应的特征值，在不违反交通规则以及正常避让路面上的障碍物的情况下，自动驾驶***可以分别从横向、纵向以及垂向三个维度生成相关的自动驾驶策略，以应对可能的路面不平情况，从而提升自动驾驶过程中的安全性和舒适性，减小车辆因振动所导致的结构疲劳损伤。
图5示出根据本申请一实施例的确定自动驾驶策略的示意图。如图5所示，在横向的维度上，可以通过换道以及道内避行的策略应对路面不平的情况。
其中，可以根据不同车道对应的点云数据的线特征，确定在对应路面最平坦的车道行驶(即换道)，例如，在线特征的特征值是沿车道方向上的一列或多列的连续多个网格中、所有点的方差的情况下，可以选择对应特征值最小的车道进行行驶。
在确定行驶的车道后，可以根据该车道上的点特征，确定车道上是否存在虚拟障碍物(如上述对应点特征的值超过预定阈值的一个网格或者一组网格的范围的虚拟障碍物)，在存在虚拟障碍物的情况下，使得车辆在车道内绕行该虚拟障碍物(即道内避行)，例如，可以采用优化的方法，在规划道内路径时，在不存在其他真实障碍物(如其他车辆、行人、锥桶等)的情况下，以路径与虚拟障碍物的距离作为约束，使得生成的路径尽可能的远离虚拟障碍物，在存在其他真实障碍物的情况下，优先避让真实障碍物。本申请还可以使用其他方法进行道内避行。
如图5所示，在纵向的维度上，根据上述路径规划得到车辆对应的路径后，可以根据点特征、线特征、面特征，确定路径上是否存在路面不平的情况，例如在对应的点特征/线特征/面特征的值超过预定阈值时，可以确定路径上对应的路面存在路面不平的情况，并根据点特征、线特征、面特征和车辆的在路面不平情况下的行驶速度范围，确定车辆当前的限速，以作为车辆在自动驾驶过程中速度的约束。
确定车辆当前限速的方法可以如公式(1)所示：
speed limit＝max((1-∝)*speed max,speed min)
其中，∝＝max(T 1,T 2,T 3)，0<∝<1      公式(1)
其中，speed limit可以表示车辆当前的限速，speed max和speed min可以预先设定，分别表示车辆在路面不平的情况下，行驶速度范围中的上限值和下限值，T 1，T 2，T 3可以分别表示归一化后的点特征值、线特征值和面特征值，T 1、T 2、T 3的取值范围为0-1，T 1、T 2、T 3的最大值越大，即∝越大，可以表示路面的起伏程度越大，1-∝越小，使得(1-∝)*speed max,越小，如果(1-∝)*speed max大于speed min，得到的speed limit为(1-∝)*speed max，其值随路面起伏程度越大(即∝越大)而越小，如果(1-∝)*speed max小于speed min，得到的speed limit为speed min。由此，可以进一步减小在路面不平的情况下，车身的振动。
需要说明的是，本申请对于确定车辆当前限速的方式不作限制，只要使得路面的起伏程 度越大，对应的车辆当前限速的上限值越小即可，例如，在公式(1)中，也可以仅利用T 1、T 2、T 3中的一个或利用T 1、T 2、T 3中的任两个来确定∝的值。
如图5所示，在垂向的维度上，可以根据惯性测量单元IMU获取的垂向加速度，确定车辆是否处于路面不平的情况，在确定车辆处于路面不平的情况下，可以对车辆主动悬架的参数，例如悬架运动学K(kinematic)特性和悬架弹性运动学C(compliance)特性进行调整，以缓解在车辆在路面不平的情况下的垂向振动。例如，在垂向加速度的变化率超过预定阈值时，可以确定车辆处于路面不平的情况(如经过颠簸路段)，则可以通过调整主动悬架的刚度和阻尼以调整主动悬架的K特性和C特性，从而抑制车身的运动，保持车身的稳定，本申请对于调整动悬架的刚度和阻尼的方法不作限定。
在一种可能的实现方式中，还可以根据点特征、线特征、面特征，确定车辆是否处于路面不平的情况，例如，在点特征、线特征或面特征超过相应的阈值时，确定车辆处于路面不平的情况，从而对车辆主动悬架的K特性和C特性进行调整。
需要说明的是，自动驾驶***可以同时确定横向、纵向和垂向维度上是否存在路面不平的情况，并执行上述维度上的相关策略，也可以仅确定并执行其中的一种或两种；横向、纵向和垂向维度上还可以存在优先级关系，例如横向维度上的优先级最高，垂向维度上的优先级最低，自动驾驶***可以维度间的优先级顺次确定对应维度上是否存在路面不平的情况，并执行维度上的相关策略。
步骤S304，自动驾驶***根据生成的策略，控制车辆进行自动驾驶，并在人机界面HMI上进行显示。
根据横向维度上的策略，自动驾驶***可以对车辆在自动驾驶时的行驶路径进行规划，根据纵向维度上的策略，自动驾驶***可以对车辆在自动驾驶时的速度进行约束，根据垂向维度上的策略，自动驾驶***可以对车辆的主动悬架在自动驾驶时的K特性和C特征进行调整。
在生成相关的策略后，自动驾驶***可以通过向车辆发送控制信号以实现车辆在路面不平情况时的自动驾驶。
其中，在车辆因路面不平而采取相关策略时，可以在人机界面HMI上进行显示，以提醒驾乘人员，使得驾乘人员能够及时了解当前的路面情况和采取的自动驾驶策略，以提高驾乘人员的信任感。
图6示出根据本申请一实施例的在人机界面HMI上进行视觉图标提示的示意图。如图6所示，在车辆因路面不平而采取相关策略时，路面不平的图标可以由灰色变为高亮提示。
在一种可能的实现方式中，还可以通过在屏幕上直接显示“路面不平”的文字提示。
表1示出根据本申请一实施例的在人机界面HMI上进行语音播报的示例。
表1
优先级 自动驾驶策略 播报内容
1 横向换道 前方车道路面不平，换道绕行
2 横向车道内避让 前方车道路面不平，车道内绕行
3 纵向减速行驶 前方车道路面不平，减速慢行
其中，优先级的对应的值越小，可以表示在执行对应的自动驾驶策略时越优先进行语音播报，例如在同时需要执行横向换道和横向车道内避让的自动驾驶策略时，HMI优先语音播 报的内容是“前方车道路面不平，换道绕行”，在一种可能的实现方式中，可以在播报完后，再播报“前方车道路面不平，车道内绕行”。
上述语音播报的过程仅为一个示例，还可以采取其它方式或者其他的播报内容。
除上述视觉图标提示以及语音播报的方式之外，还可以采取其他方式在HMI上进行显示，例如可以通过播放视频的方式进行提示。
图7示出根据本申请一实施例的自动驾驶方法的流程图。本方法可用于上述车辆上，例如可通过车辆中的自动驾驶***来执行。如图7所示，该方法包括：
步骤S701，获取传感器采集的数据，所述数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据；
步骤S702，根据所述点云数据确定特征值，所述特征值用于指示车辆前方路面的起伏程度；
步骤S703，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略。
根据本申请实施例，通过获取传感器采集的车辆前方数据，根据点云数据确定特征值，可以准确地识别出当前路面不平的情况，通过根据车辆在垂直于路面方向上的加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，可以降低路面不平产生的非预期转向等对车辆精准控制的影响，提高车辆在自动驾驶过程中的安全性和舒适性，同时，可以减小车辆、车辆上的传感器等因振动产生的结构性损伤，从而提升其可靠性和寿命耐久性。
其中，传感器采集的数据，可以是车辆从传感器获取，也可以由传感器发送至服务器，从服务器获取。车辆在垂直于路面方向上的加速度数据，可以是上文所述垂向加速度数据，特征值越大，可以表示车辆前方路面的起伏程度越大。
步骤S701的示例可以参照图3中步骤S301，步骤S702的示例可以参照图3中步骤S302，步骤S703的示例可以参照图3中步骤S303。
在一种可能的实现方式中，所述特征值根据所述点云数据在垂直于路面方向上的坐标值确定。
点云数据在垂直于路面方向上的坐标值可以反映路面的起伏程度，根据本申请实施例，可以使得特征值可以更好的衡量路面不平的情况。
垂直于路面方向上的坐标值，可以例如上文所述Z方向上的值。
在一种可能的实现方式中，所述特征值根据处理后的点云数据中最高点的z值、最低点的z值、最高点和最低点的z值差、z值的平均值、z值的方差中的一种或多种确定，所述z值表示点云数据在垂直于路面方向上的坐标值。
根据本申请实施例，可以实现根据需要灵活选择确定特征值的方式，使得特征值所指示的路面起伏程度可以更加符合车辆在自动驾驶时的实际情况。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：根据所述点云数据，确定第一特征值、第二特征值和第三特征值中的至少一种，所述第一特征值用于指示所述车辆前方路面中、各车道内局部区域的起伏程度，所述第二特征值用于指示所述车辆前方路面中、各车道路面的起伏程度，所述第三特征值用于指示所述车辆前方路面中、全部车道路面的起伏程度。
根据本申请实施例，通过确定第一特征值、第二特征值和第三特征值中的至少一种，且第一特征值、第二特征值和第三特征值分别从点、线、面三个角度指示路面的起伏程度，可 以更精准全面的体现路面不平的情况，从而为后续确定自动驾驶策略提供了针对性的依据。
其中，第一特征值可以是上文所述点特征的值，第二特征值可以是上文所述线特征的值，第三特征值可以是上文所述面特征的值。其中局部区域可以包括车道内任意大小的区域，本申请对于局部区域、各车道、以及全部路面的大小不作限定，对于确定第一特征值对应的各车道路面、第二特征值对应的各局部区域以及第三特征值对应的全部车道路面的范围的方式也不作限定。
图8示出根据本申请一实施例的自动驾驶方法的流程图。如图8所示，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：
步骤S801，根据所述第二特征值，确定对应第二特征值最小的目标车道；
步骤S802，确定所述车辆切换至所述目标车道的策略。
根据本申请实施例，通过根据第二特征值，确定对应第二特征值最小的目标车道，确定车辆切换至目标车道的策略，使得车辆可以规避起伏程度大的车道，以避免路面不平的情况，从而提高车辆的安全性和舒适性，减小相关结构性损伤。
其中，第二特征值最小，可以表示对应车道的起伏程度最小，目标车道可以是上文所述感兴趣区域内的车道。
步骤S801-S802的示例可参照上文图3中步骤S303中、有关从横向维度生成自动驾驶策略的叙述。
图9示出根据本申请一实施例的自动驾驶方法的流程图。如图9所示，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：
步骤S901，根据所述第一特征值，确定对应第一特征值超过阈值的目标局部区域；
步骤S902，确定所述车辆避行所述目标局部区域的策略。
根据本申请实施例，通过根据第一特征值，确定对应第一特征值超过阈值的目标局部区域，确定车辆避行目标局部策略的依据，使得车辆可以绕开车道内的坑洼等局部不平的路面，以避免路面不平的情况，从而提高车辆的安全性和舒适性，减小相关结构性损伤。
其中，阈值可以根据需要预先设定，第一特征值超过阈值，可以表示目标局部区域的起伏程度大于预设范围，目标局部区域可以例如对应上文所述虚拟障碍物的区域，本申请对于车辆避行目标局部区域的方式不作限定，例如可以参照上文所述以规划路径与虚拟障碍物的距离作为约束，进行优化的方式。
步骤S901-S902的示例可参照上文图3中步骤S303中、有关从横向维度生成自动驾驶策略的叙述。
在一种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，其中，车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关。
根据本申请实施例，通过根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，可以实现车辆在遇到不平路面前提前减速，通过在确定限制车辆行驶速度的策略时，使得车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关，可以实现在路面起伏程度越大的情况下，车辆的限速的上限值越小，从而可以进一步减小车辆的振动幅度，提高车辆的安全性和舒适 性，减小相关结构性损伤。
其中，本申请对于确定车辆被限制的行驶速度的上限值的方式不作限定，例如可以参照上文公式(1)。
上述示例可参照上文图3中步骤S303中、有关从纵向维度生成自动驾驶策略的叙述。
在一种可能的实现方式中，根据所述加速度数据和/或所述特征值，生成用于指示所述车辆进行自动驾驶的策略，包括：在所述特征值超过阈值，或所述加速度数据的变化率超过阈值的情况下，确定用于指示对所述车辆的主动悬架的参数进行调整的策略。
根据本申请实施例，通过在特征值超过阈值，或加速度数据的变化率超过阈值的情况下，确定用于指示对车辆的主动悬架的参数进行调整的策略，可以实现在车辆经过不平路面前，提前对主动悬架的参数进行调整，从而可以减小车辆在经过不平路面时的振动幅度，提高车辆的安全性和舒适性，减小相关结构性损伤。
其中，特征值超过阈值，可以是第一特征值、第二特征值、第三特征值中的至少一种超过对应的阈值。本申请对于对车辆的主动悬架的参数进行调整的方式不作限定。
上述示例可参照上文图3中步骤S303中、有关从垂向维度生成自动驾驶策略的叙述。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：在所述加速度数据的变化率超过预定阈值的情况下，对确定的所述特征值进行调整，得到调整后的特征值，所述调整后的特征值大于调整前的特征值。
根据本申请实施例，通过在加速度数据的变化率超过阈值的情况下，对确定的特征值进行调整，可以使得调整后的特征值更真实的反映当前路面不平的情况，使得可以更具调整后的特征值生成更加精准的自动驾驶的策略，从而进一步提高车辆在自动驾驶过程中的安全性和舒适性，减小相关结构性损伤。
其中，本申请对于对确定的特征值进行调整的方式不作限定。对确定的所述特征值进行调整，可以是对第一特征值、第二特征值、第三特征值中的至少一种进行调整。
图10示出根据本申请一实施例的自动驾驶方法的流程图。如图10所示，根据所述点云数据确定特征值，包括：
步骤S1001，将车辆前方路面中的感兴趣区域划分为网格；
步骤S1002，删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，得到处理后的点云数据；
步骤S1003，根据至少一个所述网格中的处理后的点云数据，确定所述特征值。
根据本申请实施例，通过将车辆前方路面中的感兴趣区域划分为网格，删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，可以减小其他车辆、行人等障碍物对于判断路面不平度的影响，通过根据至少一个所述网格中的处理后的点云数据，确定特征值，可以实现特征值更加准确的反映路面的不平情况，从而可以更好地辅助生成相关的自动驾驶的策略。
其中，本申请对于划分的网格的大小不作限定，例如可以根据点云数据的密度确定，障碍物可以是感兴趣区域内静止或移动的物体，例如路面上的其他车辆、行人、锥桶、红绿灯等等。
步骤S1001-步骤S1003的示例，可以参照上文图3中步骤S302中的相关叙述。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：根据任一车道内的一 个或一组网格中的点云数据在垂直于路面方向上的坐标值，确定所述一个或一组网格数据对应的局部区域的特征值，得到所述第一特征值；和/或根据任一车道方向上的一列或多列网格中的点云数据在垂直于路面方向上的坐标值，确定所述任一车道对应的特征值，得到所述第二特征值；和/或根据所有网格中的点云数据在垂直于路面方向上的坐标值，确定所述第三特征值。
根据本申请实施例，通过根据相关网格中的点云数据在垂直于路面方向上的坐标值，确定第一特征值、第二特征值、第三特征值，可以更加精准地从点、线、面的角度衡量路面不平的情况。
在一种可能的实现方式中，该方法还包括：根据所述策略，生成对所述车辆前方路面的起伏程度和/或所述车辆的自动驾驶策略的提示信息，所述提示信息包括文字、图像、音频、视频中的至少一种。
根据本申请实施例，通过生成提示信息，可以使得驾乘人员能过及时了解当前路面的情况、以及车辆采取的相关策略，从而可以提高驾乘人员的信任感。
其中，本申请对于生成提示信息的方式不作限定，例如可以通过除上述文字、图像、视频、音频以外的方式进行提示，本申请对于提示信息包括的内容也不限定。
上述过程的示例可参照图3中步骤S304。
图11示出根据本申请一实施例的自动驾驶装置的结构图。如图11所示，该装置1100包括：
获取模块1101，用于获取传感器采集的数据，所述数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据；
第一确定模块1102，用于根据所述点云数据确定特征值，所述特征值用于指示车辆前方路面的起伏程度；
第二确定模块1103，用于至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略。
根据本申请实施例，通过获取传感器采集的车辆前方数据，根据点云数据确定特征值，可以准确地识别出当前路面不平的情况，通过根据车辆在垂直于路面方向上的加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，可以降低路面不平产生的非预期转向等对车辆精准控制的影响，提高车辆在自动驾驶过程中的安全性和舒适性，同时，可以减小车辆、车辆上的传感器等因振动产生的结构性损伤，从而提升其可靠性和寿命耐久性。
在一种可能的实现方式中，特征值根据所述点云数据在垂直于路面方向上的坐标值确定。
点云数据在垂直于路面方向上的坐标值可以反映路面的起伏程度，根据本申请实施例，可以使得特征值可以更好的衡量路面不平的情况。
在一种可能的实现方式中，所述特征值根据处理后的点云数据中最高点的z值、最低点的z值、最高点和最低点的z值差、z值的平均值、z值的方差中的一种或多种确定，所述z值表示点云数据在垂直于路面方向上的坐标值。
根据本申请实施例，可以实现根据需要灵活选择确定特征值的方式，使得特征值所指示的路面起伏程度可以更加符合车辆在自动驾驶时的实际情况。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：根据所述点云数据，确定第一特征值、第二特征值和第三特征值中的至少一种，所述第一特征值用于指示所述车 辆前方路面中、各车道内局部区域的起伏程度，所述第二特征值用于指示所述车辆前方路面中、各车道路面的起伏程度，所述第三特征值用于指示所述车辆前方路面中、全部车道路面的起伏程度。
根据本申请实施例，通过确定第一特征值、第二特征值和第三特征值中的至少一种，且第一特征值、第二特征值和第三特征值分别从点、线、面三个角度指示路面的起伏程度，可以更精准全面的体现路面不平的情况，从而为后续确定自动驾驶策略提供了针对性的依据。
在一种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据所述第二特征值，确定对应第二特征值最小的目标车道；确定所述车辆切换至所述目标车道的策略。
根据本申请实施例，通过根据第二特征值，确定对应第二特征值最小的目标车道，确定车辆切换至目标车道的策略，使得车辆可以规避起伏程度大的车道，以避免路面不平的情况，从而提高车辆的安全性和舒适性，减小相关结构性损伤。
在一种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据所述第一特征值，确定对应第一特征值超过阈值的目标局部区域；确定所述车辆避行所述目标局部区域的策略。
根据本申请实施例，通过根据第一特征值，确定对应第一特征值超过阈值的目标局部区域，确定车辆避行目标局部策略的依据，使得车辆可以绕开车道内的坑洼等局部不平的路面，以避免路面不平的情况，从而提高车辆的安全性和舒适性，减小相关结构性损伤。
在一种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，其中，车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关。
根据本申请实施例，通过根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，可以实现车辆在遇到不平路面前提前减速，通过在确定限制车辆行驶速度的策略时，使得车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关，可以实现在路面起伏程度越大的情况下，车辆的限速的上限值越小，从而可以进一步减小车辆的振动幅度，提高车辆的安全性和舒适性，减小相关结构性损伤。
在一种可能的实现方式中，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：在所述特征值超过阈值，或所述加速度数据的变化率超过阈值的情况下，确定对所述车辆的主动悬架的参数进行调整的策略。
根据本申请实施例，通过在特征值超过阈值，或加速度数据的变化率超过阈值的情况下，确定用于指示对车辆的主动悬架的参数进行调整的策略，可以实现在车辆经过不平路面前，提前对主动悬架的参数进行调整，从而可以减小车辆在经过不平路面时的振动幅度，提高车辆的安全性和舒适性，减小相关结构性损伤。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：在所述加速度数据的变化率超过预定阈值的情况下，对确定的所述特征值进行调整，得到调整后的特征值，所述调整后的特征值大于调整前的特征值。
根据本申请实施例，通过在加速度数据的变化率超过阈值的情况下，对确定的特征值进 行调整，可以使得调整后的特征值更真实的反映当前路面不平的情况，使得可以更具调整后的特征值生成更加精准的自动驾驶的策略，从而进一步提高车辆在自动驾驶过程中的安全性和舒适性，减小相关结构性损伤。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：将车辆前方路面中的感兴趣区域划分为网格；删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，得到处理后的点云数据；根据至少一个所述网格中的处理后的点云数据，确定所述特征值。
根据本申请实施例，通过将车辆前方路面中的感兴趣区域划分为网格，删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，可以减小其他车辆、行人等障碍物对于判断路面不平度的影响，通过根据至少一个所述网格中的处理后的点云数据，确定特征值，可以实现特征值更加准确的反映路面的不平情况，从而可以更好地辅助生成相关的自动驾驶的策略。
在一种可能的实现方式中，根据所述点云数据确定特征值，包括：根据任一车道内的一个或一组网格中的点云数据在垂直于路面方向上的坐标值，确定所述一个或一组网格数据对应的局部区域的特征值，得到所述第一特征值；和/或根据任一车道方向上的一列或多列网格中的点云数据在垂直于路面方向上的坐标值，确定所述任一车道对应的特征值，得到所述第二特征值；和/或根据所有网格中的点云数据在垂直于路面方向上的坐标值，确定所述第三特征值。
根据本申请实施例，通过根据相关网格中的点云数据在垂直于路面方向上的坐标值，确定第一特征值、第二特征值、第三特征值，可以更加精准地从点、线、面的角度衡量路面不平的情况。
在一种可能的实现方式中，该装置还包括：生成模块，用于根据所述策略，生成对所述车辆前方路面的起伏程度和/或所述车辆的自动驾驶策略的提示信息，所述提示信息包括文字、图像、音频、视频中的至少一种。
根据本申请实施例，通过生成提示信息，可以使得驾乘人员能过及时了解当前路面的情况、以及车辆采取的相关策略，从而可以提高驾乘人员的信任感。
图12示出根据本申请一实施例的自动驾驶装置的结构图。该自动驾驶装置可适用于图1示出的车辆中，执行上述图3-图10中任一项所示出的自动驾驶方法中的功能。例如，该自动驾驶装置可以是上述车辆，也可以是可设置于该车辆内部的芯片(***)或其他部件或组件。又例如，该自动驾驶装置也可以是上述自动驾驶装置1100。
如图12所示，自动驾驶装置1200可以包括处理器1201和收发器1202。可选地，自动驾驶装置1200可以包括存储器1203。其中，处理器1201与收发器1202和存储器1203耦合，如可以通过通信总线连接。
下面结合图12对自动驾驶装置1200的各个构成部件进行具体的介绍。
上述处理器1201是自动驾驶装置1200的控制中心，可以是一个处理器，也可以是多个处理元件的统称。例如，处理器1201是一个或多个中央处理器(central processing unit，CPU)，也可以是特定集成电路(application specific integrated circuit，ASIC)，或者是被配置成实施本申请实施例的一个或多个集成电路，例如：一个或多个微处理器(digital signal processor，DSP)，或，一个或者多个现场可编程门阵列(field programmable gate array，FPGA)。
可选地，处理器1201可以通过运行或执行存储在存储器1203内的软件程序，以及调用存储在存储器1203内的数据，执行自动驾驶装置1200的各种功能。
在具体的实现中，作为一种实施例，处理器1201可以包括一个或多个CPU，例如图12中所示出的CPU0和CPU1。
在一种可能的实现方式中，自动驾驶装置1200也可以包括多个处理器，例如图12中所示的处理器1201和处理器1204。这些处理器中的每一个可以是一个单核处理器(single-CPU)，也可以是一个多核处理器(multi-CPU)。这里的处理器可以指一个或多个通信设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
上述收发器1202，用于与其他装置之间的通信。例如，参考图2，自动驾驶装置1200为车辆，收发器1202可以用于与服务器通信。
可选地，收发器1202可以包括接收器和发送器(图12中未单独示出)。其中，接收器用于实现接收功能，发送器用于实现发送功能。
可选地，收发器1202可以和处理器1201集成在一起，也可以独立存在，并通过自动驾驶装置1200的输入/输出端口(图12中未示出)与处理器1201耦合。
上述存储器1203可用于存储执行本申请方案的软件程序，并由处理器1201来控制执行，具体实现方式可以参考上述方法实施例，此处不再赘述。
其中，存储器1203可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储通信设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储通信设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储通信设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。需要说明的是，存储器1203可以和处理器1201集成在一起，也可以独立存在，并通过自动驾驶装置1200的输入/输出端口(图12中未示出)与处理器1201耦合。
需要说明的是，图12中所示出的自动驾驶装置1200的结构并不构成对自动驾驶装置的实现方式的限定，实际的自动驾驶装置可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
本申请的实施例提供了一种自动驾驶装置，包括：处理器以及用于存储处理器可执行指令的存储器；其中，所述处理器被配置为执行所述指令时实现上述方法。
本申请的实施例提供了一种终端设备，该终端设备可以执行上述方法。
本申请的实施例提供了一种非易失性计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述方法。
本申请的实施例提供了一种计算机程序产品，包括计算机可读代码，或者承载有计算机可读代码的非易失性计算机可读存储介质，当所述计算机可读代码在电子设备的处理器中运行时，所述电子设备中的处理器执行上述方法。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是――但不限于――电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的 例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(Random Access Memory，RAM)、只读存储器(Read Only Memory，ROM)、可擦式可编程只读存储器(Electrically Programmable Read-Only-Memory，EPROM或闪存)、静态随机存取存储器(Static Random-Access Memory，SRAM)、便携式压缩盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、数字多功能盘(Digital Video Disc，DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。
这里所描述的计算机可读程序指令或代码可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本申请操作的计算机程序指令可以是汇编指令、指令集架构(Instruction Set Architecture，ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(Local Area Network，LAN)或广域网(Wide Area Network，WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可编程逻辑电路、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或可编程逻辑阵列(Programmable Logic Array，PLA)，该电子电路可以执行计算机可读程序指令，从而实现本申请的各个方面。
这里参照根据本申请实施例的方法、装置(***)和计算机程序产品的流程图和/或框图描述了本申请的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其它可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
也可以把计算机可读程序指令加载到计算机、其它可编程数据处理装置、或其它设备上，使得在计算机、其它可编程数据处理装置或其它设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其它可编程数据处理装置、或其它设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本申请的多个实施例的装置、***、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，所述模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。
也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行相应的功能或动作的硬件(例如电路或ASIC(Application Specific Integrated Circuit，专用集成电路))来实现，或者可以用硬件和软件的组合，如固件等来实现。
尽管在此结合各实施例对本发明进行了描述，然而，在实施所要求保护的本发明过程中，本领域技术人员通过查看所述附图、公开内容、以及所附权利要求书，可理解并实现所述公开实施例的其它变化。在权利要求中，“包括”(comprising)一词不排除其他组成部分或步骤，“一”或“一个”不排除多个的情况。单个处理器或其它单元可以实现权利要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施，但这并不表示这些措施不能组合起来产生良好的效果。
以上已经描述了本申请的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术的改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。

Claims (16)

  1. 一种自动驾驶方法，其特征在于，所述方法包括：
    获取传感器采集的数据，所述数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据；
    根据所述点云数据确定特征值，所述特征值用于指示车辆前方路面的起伏程度；
    至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略。
  2. 根据权利要求1所述的方法，其特征在于，所述特征值根据所述点云数据在垂直于路面方向上的坐标值确定。
  3. 根据权利要求1所述的方法，其特征在于，根据所述点云数据确定特征值，包括：
    根据所述点云数据，确定第一特征值、第二特征值和第三特征值中的至少一种，所述第一特征值用于指示所述车辆前方路面中、各车道内局部区域的起伏程度，所述第二特征值用于指示所述车辆前方路面中、各车道路面的起伏程度，所述第三特征值用于指示所述车辆前方路面中、全部车道路面的起伏程度。
  4. 根据权利要求3所述的方法，其特征在于，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：
    根据所述第二特征值，确定对应第二特征值最小的目标车道；
    确定所述车辆切换至所述目标车道的策略。
  5. 根据权利要求3或4所述的方法，其特征在于，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：
    根据所述第一特征值，确定对应第一特征值超过阈值的目标局部区域；
    确定所述车辆避行所述目标局部区域的策略。
  6. 根据权利要求3-5任一项所述的方法，其特征在于，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：
    根据第一特征值、第二特征值、第三特征值中的至少一种，确定限制所述车辆行驶速度的策略，其中，车辆被限制的行驶速度的上限值，与第一特征值、第二特征值、第三特征值中的至少一种中的最大值负相关。
  7. 根据权利要求1-6任一项所述的方法，其特征在于，至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略，包括：
    在所述特征值超过阈值，或所述加速度数据的变化率超过阈值的情况下，确定对所述车辆的主动悬架的参数进行调整的策略。
  8. 根据权利要求1所述的方法，其特征在于，根据所述点云数据确定特征值，包括：
    在所述加速度数据的变化率超过预定阈值的情况下，对确定的所述特征值进行调整，得到调整后的特征值，所述调整后的特征值大于调整前的特征值。
  9. 根据权利要求1-8任一项所述的方法，其特征在于，根据所述点云数据确定特征值，包括：
    将车辆前方路面中的感兴趣区域划分为网格；
    删除所述感兴趣区域内的点云数据中，与障碍物对应的点云数据，得到处理后的点云数据；
    根据至少一个所述网格中的处理后的点云数据，确定所述特征值。
  10. 根据权利要求9所述的方法，其特征在于，根据所述点云数据确定特征值，包括：
    根据任一车道内的一个或一组网格中的点云数据在垂直于路面方向上的坐标值，确定所述一个或一组网格数据对应的局部区域的特征值，得到所述第一特征值；和/或
    根据任一车道方向上的一列或多列网格中的点云数据在垂直于路面方向上的坐标值，确定所述任一车道对应的特征值，得到所述第二特征值；和/或
    根据所有网格中的点云数据在垂直于路面方向上的坐标值，确定所述第三特征值。
  11. 根据权利要求1-10任一项所述的方法，其特征在于，所述特征值根据处理后的点云数据中最高点的z值、最低点的z值、最高点和最低点的z值差、z值的平均值、z值的方差中的一种或多种确定，所述z值表示点云数据在垂直于路面方向上的坐标值。
  12. 根据权利要求1-11中任一项所述的方法，其特征在于，所述方法还包括：
    根据所述策略，生成对所述车辆前方路面的起伏程度和/或所述车辆的自动驾驶策略的提示信息，所述提示信息包括文字、图像、音频、视频中的至少一种。
  13. 一种自动驾驶装置，其特征在于，所述装置包括：
    获取模块，用于获取传感器采集的数据，所述数据包括车辆前方的点云数据和车辆在垂直于路面方向上的加速度数据；
    第一确定模块，用于根据所述点云数据确定特征值，所述特征值用于指示车辆前方路面的起伏程度；
    第二确定模块，用于至少根据所述加速度数据和/或所述特征值，确定所述车辆的自动驾驶策略。
  14. 一种自动驾驶装置，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为执行所述指令时实现权利要求1-12任意一项所述的方法。
  15. 一种非易失性计算机可读存储介质，其上存储有计算机程序指令，其特征在于，所述计算机程序指令被处理器执行时实现权利要求1-12中任意一项所述的方法。
  16. 一种计算机程序产品，包括计算机可读代码，或者承载有计算机可读代码的非易失性计算机可读存储介质，当所述计算机可读代码在电子设备中运行时，所述电子设备中的处理器执行权利要求1-12中任意一项所述的方法。
PCT/CN2021/105476 2021-07-09 2021-07-09 自动驾驶方法、装置和存储介质 WO2023279371A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202180006112.6A CN114616158A (zh) 2021-07-09 2021-07-09 自动驾驶方法、装置和存储介质
PCT/CN2021/105476 WO2023279371A1 (zh) 2021-07-09 2021-07-09 自动驾驶方法、装置和存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/105476 WO2023279371A1 (zh) 2021-07-09 2021-07-09 自动驾驶方法、装置和存储介质

Publications (1)

Publication Number Publication Date
WO2023279371A1 true WO2023279371A1 (zh) 2023-01-12

Family

ID=81857213

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/105476 WO2023279371A1 (zh) 2021-07-09 2021-07-09 自动驾驶方法、装置和存储介质

Country Status (2)

Country Link
CN (1) CN114616158A (zh)
WO (1) WO2023279371A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115871622A (zh) * 2023-01-19 2023-03-31 重庆赛力斯新能源汽车设计院有限公司 基于落差路面的辅助驾驶方法、电子设备及存储介质
CN115909787A (zh) * 2023-02-06 2023-04-04 山东科技大学 一种路面破损情况的预警方法、设备及介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118269997A (zh) * 2022-12-30 2024-07-02 华为技术有限公司 一种路面检测方法及相关装置

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018007079A1 (en) * 2016-07-08 2018-01-11 Jaguar Land Rover Limited Improvements in vehicle speed control
CN108459319A (zh) * 2018-03-13 2018-08-28 燕山大学 一种车辆行驶区域地形高度快速扫描***
CN109635672A (zh) * 2018-11-22 2019-04-16 同济大学 一种无人驾驶的道路特征参数估计方法
US10486485B1 (en) * 2017-04-19 2019-11-26 Zoox, Inc. Perception based suspension control
CN111339229A (zh) * 2020-02-24 2020-06-26 交通运输部水运科学研究所 一种船舶自主航行辅助决策***
CN112141111A (zh) * 2020-09-02 2020-12-29 新石器慧义知行智驰(北京)科技有限公司 无人车行驶控制方法、装置及无人车
US20210096570A1 (en) * 2019-09-30 2021-04-01 Gm Cruise Holdings Llc Road quality based routing

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108955584B (zh) * 2017-05-23 2021-02-05 上海汽车集团股份有限公司 一种路面探测方法及装置
CN108221603B (zh) * 2018-01-08 2019-08-09 重庆大学 一种道路的路面三维信息检测装置、方法及***
CN109002039B (zh) * 2018-07-06 2021-04-06 达闼科技（北京）有限公司 避障提醒方法、相关装置及计算机可读存储介质
CN109683170B (zh) * 2018-12-27 2021-07-02 驭势科技（北京）有限公司 一种图像行驶区域标注方法、装置、车载设备及存储介质
CN110516564A (zh) * 2019-08-06 2019-11-29 深兰科技（上海）有限公司 路面检测方法和装置
CN112308052B (zh) * 2020-12-29 2021-08-17 智道网联科技(北京)有限公司 一种道路异常区域检测方法、装置、电子设备及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018007079A1 (en) * 2016-07-08 2018-01-11 Jaguar Land Rover Limited Improvements in vehicle speed control
US10486485B1 (en) * 2017-04-19 2019-11-26 Zoox, Inc. Perception based suspension control
CN108459319A (zh) * 2018-03-13 2018-08-28 燕山大学 一种车辆行驶区域地形高度快速扫描***
CN109635672A (zh) * 2018-11-22 2019-04-16 同济大学 一种无人驾驶的道路特征参数估计方法
US20210096570A1 (en) * 2019-09-30 2021-04-01 Gm Cruise Holdings Llc Road quality based routing
CN111339229A (zh) * 2020-02-24 2020-06-26 交通运输部水运科学研究所 一种船舶自主航行辅助决策***
CN112141111A (zh) * 2020-09-02 2020-12-29 新石器慧义知行智驰(北京)科技有限公司 无人车行驶控制方法、装置及无人车

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115871622A (zh) * 2023-01-19 2023-03-31 重庆赛力斯新能源汽车设计院有限公司 基于落差路面的辅助驾驶方法、电子设备及存储介质
CN115909787A (zh) * 2023-02-06 2023-04-04 山东科技大学 一种路面破损情况的预警方法、设备及介质

Also Published As

Publication number Publication date
CN114616158A (zh) 2022-06-10

Similar Documents

Publication Publication Date Title
WO2023279371A1 (zh) 自动驾驶方法、装置和存储介质
US10551840B2 (en) Planning driven perception system for autonomous driving vehicles
US11238733B2 (en) Group driving style learning framework for autonomous vehicles
US10520319B2 (en) Data driven map updating system for autonomous driving vehicles
US11378956B2 (en) Perception and planning collaboration framework for autonomous driving
EP3500965B1 (en) Speed control for a full stop of an autonomous driving vehicle
US10775801B2 (en) Determining speeds along a path for autonomous driving vehicles
WO2019169604A1 (en) Simulation-based method to evaluate perception requirement for autonomous driving vehicles
US20200361485A1 (en) A pedestrian interaction system for low speed scenes for autonomous vehicles
US10691129B2 (en) Dynamically adjustable reference line sampling point density for autonomous vehicles
US10053091B2 (en) Spring system-based change lane approach for autonomous vehicles
US20190196485A1 (en) Non-blocking boundary for autonomous vehicle planning
US10569651B2 (en) Speed control and steering control assistant based on pitch status and roll status of autonomous driving vehicle
US10983522B2 (en) Emergency stop speed profile for autonomous vehicles
US10409280B2 (en) Control dominated planning and control system for autonomous driving vehicles
US11260880B2 (en) Map-less and localization-less lane following method for autonomous driving of autonomous driving vehicles on highway
US20190325234A1 (en) Lane post-processing in an autonomous driving vehicle
WO2020132959A1 (en) Spiral curve based vertical parking planner system for autonomous driving vehicles
US11400955B2 (en) Multi-point enforced based stitch method to connect two smoothed reference lines
JP2019196941A (ja) 自車位置推定装置
US20230288218A1 (en) Navigation system with voice assistant mechanism and method of operation thereof

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21948865

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21948865

Country of ref document: EP

Kind code of ref document: A1