CN113978457B - Collision risk prediction method and device - Google Patents
Collision risk prediction method and device Download PDFInfo
- Publication number
- CN113978457B CN113978457B CN202111593273.8A CN202111593273A CN113978457B CN 113978457 B CN113978457 B CN 113978457B CN 202111593273 A CN202111593273 A CN 202111593273A CN 113978457 B CN113978457 B CN 113978457B
- Authority
- CN
- China
- Prior art keywords
- target object
- determining
- target
- motion
- bounding box
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000004590 computer program Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- 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
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
-
- 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
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
-
- 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
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
-
- 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
- B60W2554/00—Input parameters relating to objects
- B60W2554/40—Dynamic objects, e.g. animals, windblown objects
- B60W2554/402—Type
- B60W2554/4029—Pedestrians
-
- 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
- B60W2554/00—Input parameters relating to objects
- B60W2554/80—Spatial relation or speed relative to objects
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a collision risk prediction method and a collision risk prediction device, which relate to the technical field of safe driving, and the specific scheme comprises the following steps: the computer equipment obtains at least one target bounding box by obtaining attribute information and motion attitude information of a target object and determining the position, size and number of bounding boxes corresponding to the target object according to the attribute information of the target object, then determines the motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track, then determines the motion track of the bounding box of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track, and finally determines the collision probability of the current vehicle and the target object according to the first motion track and the second motion track. Under the condition that the measurement precision of the sensor is not changed, the method improves the prediction accuracy of the collision probability under the condition of complex road conditions.
Description
Technical Field
The invention relates to the field of safe driving, in particular to a collision risk prediction method and device.
Background
In the driving process of the vehicle, the degree of the danger of the vehicle by other vehicles or pedestrians in the road can be judged by predicting the collision probability of the vehicle and other vehicles or pedestrians in the road, so that effective reference is provided for road dangerous collision early warning or obstacle avoidance route planning.
According to the collision risk prediction method in the traditional technology, target information of other vehicles or pedestrians in a road is obtained through a vehicle-mounted sensor, the other vehicles or pedestrians in a complex shape in the road are replaced by a simple geometric surrounding body through a surrounding box in a simple shape, track prediction is respectively carried out on the basis of motion posture changes of the other vehicles or pedestrians in the own vehicle and the road, and finally the collision probability between the own vehicle and the other vehicles or pedestrians in the road is calculated on the basis of the track of the own vehicle and the track of the other vehicles or pedestrians in the road.
However, the motion attitude of the vehicle or the pedestrian is changed more complicatedly during the motion process, especially when the curvature of the road is larger, so that the accuracy of the target information measured by the vehicle-mounted sensor is reduced, and the prediction accuracy of the traditional collision risk prediction method under the condition of a complex road condition is lower.
Disclosure of Invention
The invention provides a collision risk prediction method and a collision risk prediction device, which solve the problem of low prediction accuracy of the traditional collision risk prediction method under the condition of complex road conditions.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, the present invention provides a collision risk prediction method, including:
acquiring attribute information and motion attitude information of a target object; the target object is an object whose distance from the current vehicle is less than or equal to a first threshold value;
determining the position, the size and the number of bounding boxes corresponding to the target object according to the attribute information of the target object to obtain at least one target bounding box; the target bounding box covers the target object;
determining a motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track;
determining a motion track of an enclosure of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track;
and determining the collision probability of the current vehicle and the target object according to the first motion trail and the second motion trail.
With reference to the first aspect, in a possible implementation manner, determining a collision probability between a current vehicle and a target object according to a first motion trajectory and a second motion trajectory includes: predicting the deviation of the first motion track and the deviation of the second motion track to obtain a first target motion track and a second target motion track; determining a dangerous area around the first target motion track when the first target motion track and the second target motion track are overlapped; the danger zone comprises a plurality of sub-danger zones with different danger levels; determining the overlapping probability of the second motion track and each sub-danger area to obtain a plurality of overlapping probabilities; and acquiring the minimum value of the multiple overlapping probabilities to obtain the collision probability of the current vehicle and the target object.
With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, after obtaining the collision probability of the current vehicle and the target object, the method further includes: determining a sub-danger zone corresponding to the minimum value of the overlapping probability; and determining the danger level of the collision accident between the current vehicle and the target object according to the danger level corresponding to the sub-danger area.
With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, determining a risk level of a collision accident between a current vehicle and a target object according to a risk level corresponding to a sub-risk area includes: and determining the danger level corresponding to the sub-danger area according to the width of the sub-danger area.
With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, the attribute information includes a type, a length, and a width of the target object, and the determining, according to the attribute information of the target object, a position, a size, and a number of bounding boxes corresponding to the target object to obtain at least one target bounding box includes: determining the size of a bounding box corresponding to the target object according to the width of the target object; determining the number of bounding boxes corresponding to the target object according to the length of the target object; and determining the position of the bounding box corresponding to the target object according to the type, the length and the width of the target object.
With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, determining a position of a bounding box corresponding to a target object according to a type, a length, and a width of the target object includes: determining the reference point position of the bounding box corresponding to the target object according to the type, the length and the width of the target object; and determining the position of the bounding box corresponding to the target object according to the position of the reference point.
With reference to the first aspect and the foregoing possible implementation manners, in another possible implementation manner, the determining, according to the type, the length, and the width of the target object, the reference point position of the bounding box corresponding to the target object, where the type of the target object includes an automobile, a non-automobile, and a pedestrian, includes: when the type of the target object is the motor vehicle, determining the center of a head line or the center of a tail line of the motor vehicle as the reference point position of the bounding box corresponding to the target object; when the type of the target object is a non-motor vehicle or a pedestrian, the geometric center of the non-motor vehicle or the pedestrian is determined as the reference point position of the bounding box corresponding to the target object.
In a second aspect, the present invention provides a collision risk prediction apparatus, comprising: the device comprises an acquisition module, a surrounding module, a first determination module, a second determination module and an output module.
The acquisition module is used for acquiring attribute information and motion attitude information of the target object; the target object is an object whose distance from the current vehicle is less than or equal to a first threshold value;
the bounding module is used for determining the position, the size and the number of bounding boxes corresponding to the target object according to the attribute information of the target object to obtain at least one target bounding box; the target bounding box covers the target object;
the first determining module is used for determining the motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track;
the second determining module is used for determining the motion track of the bounding box of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track;
and the output module is used for determining the collision probability of the current vehicle and the target object according to the first motion trail and the second motion trail.
With reference to the second aspect, in a possible implementation manner, the output module is specifically configured to: predicting the deviation of the first motion track and the deviation of the second motion track to obtain a first target motion track and a second target motion track; determining a dangerous area around the first target motion track when the first target motion track and the second target motion track are overlapped; the danger zone comprises a plurality of sub-danger zones with different danger levels; determining the overlapping probability of the second motion track and each sub-danger area to obtain a plurality of overlapping probabilities; and acquiring the minimum value of the multiple overlapping probabilities to obtain the collision probability of the current vehicle and the target object.
With reference to the second aspect and the foregoing possible implementation manners, in another possible implementation manner, the output module is further configured to: determining a sub-danger zone corresponding to the minimum value of the overlapping probability; and determining the danger level of the collision accident between the current vehicle and the target object according to the danger level corresponding to the sub-danger area.
With reference to the second aspect and the foregoing possible implementation manners, in another possible implementation manner, before determining the risk level of the collision accident between the current vehicle and the target object according to the risk level corresponding to the sub-risk area, the output module is further configured to: and determining the danger level corresponding to the sub-danger area according to the width of the sub-danger area.
With reference to the second aspect and the foregoing possible implementation manners, in another possible implementation manner, the enclosure module is specifically configured to: determining the size of a bounding box corresponding to the target object according to the width of the target object; determining the number of bounding boxes corresponding to the target object according to the length of the target object; and determining the position of the bounding box corresponding to the target object according to the type, the length and the width of the target object.
With reference to the second aspect and the foregoing possible implementation manners, in another possible implementation manner, the enclosure module is specifically configured to: determining the reference point position of the bounding box corresponding to the target object according to the type, the length and the width of the target object; and determining the position of the bounding box corresponding to the target object according to the position of the reference point.
With reference to the second aspect and the possible implementations described above, in another possible implementation, the types of the target object include an automobile, a non-automobile and a pedestrian, and the enclosing module is specifically configured to: when the type of the target object is the motor vehicle, determining the center of a head line or the center of a tail line of the motor vehicle as the reference point position of the bounding box corresponding to the target object; when the type of the target object is a non-motor vehicle or a pedestrian, the geometric center of the non-motor vehicle or the pedestrian is determined as the reference point position of the bounding box corresponding to the target object.
In a third aspect, the present invention provides a computer apparatus comprising: a processor and a memory. The memory is for storing computer program code, the computer program code including computer instructions. When the processor executes the computer instructions, the computer device performs a collision risk prediction method as in the first aspect and any possible implementation thereof.
In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions which, when run on a computer device, cause the computer device to perform a collision risk prediction method as defined in the first aspect or any one of the possible implementations of the first aspect.
According to the collision risk prediction method provided by the embodiment of the invention, the computer equipment obtains the attribute information and the motion attitude information of the target object, determines the position, the size and the number of the bounding boxes corresponding to the target object according to the attribute information of the target object to obtain at least one target bounding box, determines the motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track, determines the motion track of the bounding box of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track, and finally determines the collision probability between the current vehicle and the target object according to the first motion track and the second motion track. In this embodiment, different target objects may correspond to target bounding boxes of different numbers, positions, and sizes, so that the bounding modes for different target objects are more flexible and effective, and the accuracy of predicting the collision probability under the condition of a complex road condition is improved under the condition that the measurement accuracy of the sensor is not changed.
Drawings
Fig. 1 is a schematic view of an application scenario of a collision risk prediction method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a collision risk prediction method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of enclosure in the form of a multi-enclosure box according to an embodiment of the present invention;
FIG. 4 is a schematic view of a hazardous area around an enclosure provided by an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a collision risk prediction apparatus according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the following, the terms "first", "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present disclosure, "a plurality" means two or more unless otherwise specified.
Additionally, the use of "based on" or "according to" means open and inclusive, as a process, step, calculation, or other action that is "based on" or "according to" one or more stated conditions or values may in practice be based on additional conditions or exceeding the stated values.
In order to solve the problem that the prediction accuracy of a traditional collision risk prediction method is low under the condition of complex road conditions, the embodiment of the invention provides a collision risk prediction method and a device.
Therefore, different target objects can correspond to the target bounding boxes with different quantities, positions and sizes, so that the bounding mode for the target objects is more flexible and effective, and the prediction accuracy of the collision probability under the condition of complicated road conditions is improved under the condition that the measurement precision of the sensor is unchanged.
Fig. 1 is an application scenario diagram of a collision risk prediction method according to an embodiment of the present invention, where when a current vehicle 101 in fig. 1 runs in a road, attribute information and motion posture information of a target vehicle 103 or a pedestrian 104 in the road are acquired by a sensor 102, and a collision probability between the current vehicle and the target vehicle or the pedestrian is calculated by a computer device 105 by using the collision risk prediction method in the present invention. The target vehicle 103 may be a motor vehicle, including but not limited to a special vehicle or a private vehicle such as a motorcycle, a car, a minibus, a truck, a fire engine, and the like, or a non-motor vehicle, including but not limited to a tricycle, a bicycle, a body sensing vehicle, and the like, and the type of the target vehicle 103 is not limited herein.
The execution subject of the collision risk prediction method provided by the embodiment of the invention is computer equipment. The computer device may be a terminal, a server, or a server cluster. The terminal may be a vehicle-mounted terminal in the target vehicle, a CPU in the vehicle-mounted terminal, or a client of the vehicle-mounted terminal.
Based on the above description of the computer device, an embodiment of the present invention provides a collision risk prediction method, as shown in fig. 2, including the following steps S201 to S205.
S201, acquiring attribute information and motion attitude information of a target object; the target object is an object whose distance from the current vehicle is less than or equal to a first threshold value.
The attribute information of the target object may include a type, a length, and a width of the target object, the type of the target object may be a target vehicle or a pedestrian, and the target vehicle may be an automobile or a non-automobile. The motion pose information may include a current position, a motion direction, a velocity, and an acceleration of the target object.
In one possible implementation manner, the computer device may acquire attribute information and an operation posture of a target object within a first threshold range around the current vehicle through an in-vehicle sensor. The magnitude of the first threshold may be fixed or may be dynamically changed according to time or road traffic conditions, and is not limited herein.
S202, determining the position, the size and the number of bounding boxes corresponding to the target object according to the attribute information of the target object to obtain at least one target bounding box; the target bounding box covers the target object.
It is understood that, in the conventional art, bounding boxes covering different target objects are all single bounding boxes, and the reference positions of the bounding boxes are unchanged. In the bounding box generation scheme provided in this step, the computer device may determine the position, size, and number of bounding boxes corresponding to the target object according to the attribute information of the target object. Illustratively, taking the shape of the bounding box as a circle, as shown in FIG. 3, the length of the target objectlThe longer the number of bounding boxes covering the target object can be, the greater the width of the target objectwThe wider the radius of the bounding box covering the target object may be, the different types of target objects and the different locations of the bounding boxes covering the target objects.
S203, determining the motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track.
Specifically, the computer device may predict a motion trajectory of the target bounding box by using the integrator according to the current position, the motion direction, the velocity, and the acceleration of the target object, so as to obtain the first motion trajectory.
S204, determining the motion track of the bounding box of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track.
Specifically, the computer device may determine the number, size, and position of the bounding box of the current vehicle according to the attribute information of the current vehicle, and predict the motion trajectory of the bounding box of the current vehicle by using the integrator in combination with the current position, motion direction, speed, and acceleration of the current vehicle, so as to obtain the second motion trajectory.
And S205, determining the collision probability of the current vehicle and the target object according to the first motion track and the second motion track.
Specifically, the computer device may predict a time at which the first motion trajectory and the second motion trajectory are closest to each other in a future period of time, and calculate a variance of respective positions of the target bounding box and the bounding box of the current vehicle when the bounding box is closest to each other by a linear or nonlinear method, such as a kalman filter method, on the assumption that trajectory errors of the first motion trajectory and the second motion trajectory satisfy a gaussian distribution, thereby calculating a collision probability of the current vehicle and the target object.
In the collision risk prediction method in this embodiment, the computer device obtains the attribute information and the motion posture information of the target object, determines the position, the size, and the number of the bounding boxes corresponding to the target object according to the attribute information of the target object to obtain at least one target bounding box, determines the motion trajectory of the target bounding box according to the motion posture information of the target object to obtain a first motion trajectory, determines the motion trajectory of the bounding box of the current vehicle according to the motion posture information of the current vehicle and the attribute information of the current vehicle to obtain a second motion trajectory, and determines the collision probability between the current vehicle and the target object according to the first motion trajectory and the second motion trajectory. In this embodiment, different target objects may correspond to target bounding boxes of different numbers, positions, and sizes, so that the bounding manner for the target objects is more flexible and effective, and the accuracy of predicting the collision probability under the condition of a complex road condition is improved under the condition that the measurement accuracy of the sensor is not changed.
In a possible implementation manner, on the basis of the foregoing embodiment, the foregoing step S205 includes:
s301, predicting the deviation of the first motion track and the deviation of the second motion track to obtain a first target motion track and a second target motion track.
In a possible implementation manner, under the assumption that the errors of the first motion trajectory and the second motion trajectory satisfy the gaussian distribution, the computer device may predict the deviation of the first motion trajectory and the deviation of the second motion trajectory by a linear or nonlinear method to obtain a plurality of possible first target motion trajectories and second target motion trajectories.
S302, determining a dangerous area around the first target motion track when the first target motion track is overlapped with the second target motion track; the hazard zone includes a plurality of sub-hazard zones of different hazard classes.
As shown in fig. 4, the dangerous area may be a middle area formed by overlapping a certain safety distance on the edge of the target bounding box, and the width of the dangerous area is the safety distance. The safe distance is related to the road curvature, the road width and the self speed of the current position of the target object, the safe distance is dynamically changed along with the movement of the target object, the safe distances are different, and the corresponding danger levels of the danger areas can be different. It will be appreciated that the greater the safety distance, the greater the area of the hazard zone, with regions closer to the center of the hazard zone having a higher risk and regions further from the center of the hazard zone having a lower risk.
It will be appreciated that, of the plurality of possible first and second target motion trajectories, the first and second target motion trajectories may or may not overlap.
In particular, the computer device may determine a danger zone around the first target motion trajectory when the first target motion trajectory and the second target motion trajectory overlap.
S303, determining the overlapping probability of the second motion track and each sub-danger area to obtain a plurality of overlapping probabilities.
Wherein, the safety distance corresponding to different sub-dangerous areas can be different. It will be appreciated that the second motion profile may overlap each of the sub-risk zones included in the risk zone due to errors in the second motion profile.
Specifically, the computer may calculate the variance of the respective positions of the target bounding box and the bounding box of the current vehicle when the target bounding box and the bounding box of the current vehicle are closest to each other by a linear or nonlinear method, thereby determining the overlapping probability of the second motion trajectory and each sub-risk region, and obtaining a plurality of overlapping probabilities.
S304, obtaining the minimum value of the multiple overlapping probabilities to obtain the collision probability of the current vehicle and the target object.
Specifically, the computer device may determine the minimum value of the plurality of overlap probabilities in step S303 as the collision probability of the current vehicle and the target object.
In the collision risk prediction method in this embodiment, the computer device obtains the first target motion trajectory and the second target motion trajectory by determining a deviation of the first motion trajectory and a deviation of the second motion trajectory, determines a danger region around the first target motion trajectory when the first target motion trajectory and the second target motion trajectory overlap, calculates an overlap probability of the second motion trajectory and each sub-danger region to obtain a plurality of overlap probabilities, and determines a minimum value of the plurality of overlap probabilities as a collision probability of the current vehicle and the target object. According to the method and the device, the collision probability is calculated based on the track error, and the accuracy of calculating the collision probability is improved.
In a possible implementation manner, on the basis of the foregoing embodiment, after the foregoing step S304, the method further includes:
s401, determining a sub-danger area corresponding to the minimum value of the overlapping probability;
s402, determining the danger level of the collision accident between the current vehicle and the target object according to the danger level corresponding to the sub-danger area.
Specifically, the computer device may determine a sub-risk area corresponding to the minimum value of the overlap probability, determine a risk level corresponding to the sub-risk area according to the width of the sub-risk area, that is, the safety distance, and determine the risk level of the collision accident between the current vehicle and the target object according to the risk level corresponding to the sub-risk area.
For example, the computer device may divide the risk level into a plurality of levels, each level corresponding to a different safety distance range, and determine, according to the safety distance corresponding to the sub-risk area, the risk level corresponding to the safety distance range in which the safety distance is located as the risk level corresponding to the sub-risk area.
In the collision risk prediction method in this embodiment, the computer device determines the sub-risk area corresponding to the minimum value of the overlap probability, and determines the risk level of the collision accident between the current vehicle and the target object according to the risk level corresponding to the sub-risk area. Because the calculated collision probability is not the overlapping probability of the target bounding box and the bounding box of the current vehicle, but the overlapping probability of the dangerous area around the target bounding box and the bounding box of the current vehicle, the dangerous level output in the embodiment can represent the dangerous degree of the collision, and a driver or an automatic driving system can plan an obstacle avoidance route according to the dangerous level.
In a possible implementation manner, on the basis of the foregoing embodiment, the attribute information of the target object includes a type, a length, and a width of the target object, and the foregoing step S202 includes:
s501, determining the size of a bounding box corresponding to the target object according to the width of the target object;
s502, determining the number of bounding boxes corresponding to the target object according to the length of the target object;
s503, determining the position of the bounding box corresponding to the target object according to the type, the length and the width of the target object;
specifically, the computer device may determine the size of the bounding box corresponding to the target object according to the width of the target object, determine the number of the bounding boxes corresponding to the target object according to the length of the target object, and determine the position of the bounding box corresponding to the target object according to the type, the length, and the width of the target object.
It will be appreciated that when the computer device generates a bounding box based on different target object types, the reference point for the location of the bounding box is different. For example, the motor vehicle may use the center of the head line or the center of the tail line as a reference point, the non-motor vehicle or the pedestrian uses the geometric center of the non-motor vehicle or the pedestrian as a reference point, and the computer device may determine the position of the bounding box corresponding to the target object according to the position of the reference point.
In one possible implementation, taking the type of target object as a motor vehicle and the shape of the bounding box as a circle as an example, the dimensions of the bounding box may be given by the radius of the circleRTo indicate that the location of the bounding box can be in the distance from the circle to the center point of the head line or the center point of the tail lineD 1 And the distance between the centers of circlesD 2 To indicate. As described aboveR、D 1 AndD 2 the calculation formula of (a) is as follows:
wherein the content of the first and second substances,lis the length of the target object and,wis the width of the target object and,nthe number of bounding boxes corresponding to the target object.nSize and ofsIn connection with, it is to be understood thatsThe larger the size of the tube is,nthe larger the target object may be, for example, when the target object is a common car,nmay be 3.
In another possible implementation, taking the type of the target object as a non-motor vehicle or a pedestrian and the shape of the bounding box as a circle as an example, the position of the bounding box can be calculated by using the geometric center of the non-motor vehicle or the pedestrian as the center of the bounding box, the radius of the circle is calculated by the same formula (1), and the number of the bounding boxes is determined by the same method as the above implementation, for example, when the target object is a pedestrian,nmay be 1.
Further, in another possible implementation manner, for the current vehicle, the position of the bounding box of the current vehicle may be determined by taking the geometric center of the current vehicle as the center of the bounding box of the current vehicle, and taking the radius of a bounding circle in the bounding box of the current vehiclerAnd the distance between each circle in the bounding box of the current vehicledThe calculation formula of (c) may be as follows:
wherein, the first and the second end of the pipe are connected with each other,Las to the length of the current vehicle,Was to the width of the current vehicle,Nthe determination method of the number of bounding boxes of the current vehicle is also the same as the above implementation method, and is not described herein again.
In the collision risk prediction method in this embodiment, the computer device determines the size of the bounding box corresponding to the target object according to the width of the target object, determines the number of the bounding boxes corresponding to the target object according to the length of the target object, and determines the position of the bounding box corresponding to the target object according to the type, the length, and the width of the target object, so that the bounding modes for different target objects are more flexible and effective, and the prediction accuracy of the collision probability under the condition of a complex road condition is improved under the condition that the measurement accuracy of the sensor is not changed.
The foregoing has outlined rather broadly the solution provided by an embodiment of the present invention from the perspective of a computer device. It will be appreciated that the computer device, in order to implement the above-described functions, comprises corresponding hardware structures and/or software modules for performing the respective functions. Those of skill in the art will readily appreciate that the present invention can be implemented in hardware or a combination of hardware and computer software, in conjunction with the exemplary algorithm steps described in connection with the embodiments disclosed herein. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
Fig. 5 is a schematic block diagram illustrating a collision risk prediction apparatus, which may include, as shown in fig. 5: an acquisition module 51, a surrounding module 52, a first determination module 53, a second determination module 54, and an output module 55.
An obtaining module 51, configured to obtain attribute information and motion posture information of a target object; the target object is an object whose distance from the current vehicle is less than or equal to a first threshold value;
the bounding module 52 is configured to determine, according to the attribute information of the target object, a position, a size, and a number of bounding boxes corresponding to the target object, to obtain at least one target bounding box; the target bounding box covers the target object;
the first determining module 53 is configured to determine a motion trajectory of the target bounding box according to the motion posture information of the target object, so as to obtain a first motion trajectory;
the second determining module 54 is configured to determine a motion trajectory of the bounding box of the current vehicle according to the motion posture information of the current vehicle and the attribute information of the current vehicle, so as to obtain a second motion trajectory;
and the output module 55 is used for determining the collision probability of the current vehicle and the target object according to the first motion trail and the second motion trail.
Optionally, the output module 55 is specifically configured to: determining the deviation of the first motion track and the deviation of the second motion track to obtain a first target motion track and a second target motion track; determining a dangerous area around the first target motion track when the first target motion track and the second target motion track are overlapped; the danger zone comprises a plurality of sub-danger zones with different danger levels; determining the overlapping probability of the second motion track and each sub-danger area to obtain a plurality of overlapping probabilities; and acquiring the minimum value of the multiple overlapping probabilities to obtain the collision probability of the current vehicle and the target object.
Optionally, the output module 55 is further configured to: determining a sub-danger zone corresponding to the minimum value of the overlapping probability; and determining the danger level of the collision accident of the current vehicle and the target object according to the danger level corresponding to the sub-danger area.
Optionally, before determining the risk level of the collision accident between the current vehicle and the target object according to the risk levels corresponding to the sub-risk areas, the output module 55 is further configured to: and determining the danger level corresponding to the sub-danger area according to the width of the sub-danger area.
Optionally, the surrounding module 52 is specifically configured to: determining the size of a bounding box corresponding to the target object according to the width of the target object; determining the number of bounding boxes corresponding to the target object according to the length of the target object; and determining the position of the bounding box corresponding to the target object according to the type, the length and the width of the target object.
Optionally, the surrounding module 52 is specifically configured to: determining the reference point position of the bounding box corresponding to the target object according to the type, the length and the width of the target object; and determining the position of the bounding box corresponding to the target object according to the position of the reference point.
Optionally, the types of target objects include motor vehicles, non-motor vehicles and pedestrians, and the surrounding module 52 is specifically configured to: when the type of the target object is the motor vehicle, determining the center of a head line or the center of a tail line of the motor vehicle as the reference point position of the bounding box corresponding to the target object; when the type of the target object is a non-motor vehicle or a pedestrian, determining the geometric center of the non-motor vehicle or the pedestrian as the reference point position of the bounding box corresponding to the target object.
The collision risk prediction device provided by the embodiment of the invention is used for executing the collision risk prediction method, so that the technical effect same as that of the collision risk prediction method can be achieved.
The embodiment of the invention also provides computer equipment, which comprises a processor and a memory; the memory is for storing computer program code, the computer program code comprising computer instructions; when the processor executes the computer instructions, the computer device executes the collision risk prediction method provided by the foregoing embodiments of the present invention.
The computer device provided by the embodiment of the invention is used for executing the collision risk prediction method, so that the technical effect same as that of the collision risk prediction method can be achieved.
Embodiments of the present invention further provide a computer storage medium, which stores one or more computer instructions and is configured to, when executed, implement the collision risk prediction method provided in the foregoing embodiments of the present invention.
The computer storage medium provided by the embodiment of the invention is used for executing the collision risk prediction method, so that the technical effect same as that of the collision risk prediction method can be achieved.
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions within the technical scope of the present invention are intended to be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (9)
1. A collision risk prediction method, comprising:
acquiring attribute information and motion attitude information of a target object; the target object is an object whose distance from the current vehicle is less than or equal to a first threshold; the attribute information comprises the type, length and width of the target object;
determining the size of a target bounding box corresponding to the target object according to the width of the target object; determining the number of the target bounding boxes corresponding to the target object according to the length of the target object; determining the position of the target bounding box corresponding to the target object according to the type, the length and the width of the target object; the target bounding box covers the target object;
determining a motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track;
determining a motion track of an enclosure of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track;
and determining the collision probability of the current vehicle and the target object according to the first motion trail and the second motion trail.
2. The method of claim 1, wherein determining the probability of collision of the current vehicle and the target object from the first motion profile and the second motion profile comprises:
predicting the deviation of the first motion track and the deviation of the second motion track to obtain a first target motion track and a second target motion track;
determining a danger zone around the first target motion trajectory when the first target motion trajectory and the second target motion trajectory overlap; the danger zone comprises a plurality of sub-danger zones of different danger levels;
determining the overlapping probability of the second motion track and each sub-danger area to obtain a plurality of overlapping probabilities;
and acquiring the minimum value of the multiple overlapping probabilities to obtain the collision probability of the current vehicle and the target object.
3. The method of claim 2, further comprising, after obtaining the probability of collision of the current vehicle with the target object:
determining a sub-danger zone corresponding to the minimum value of the overlapping probability;
and determining the danger level of the collision accident between the current vehicle and the target object according to the danger level corresponding to the sub-danger area.
4. The method according to claim 3, wherein the determining the risk level of the collision accident between the current vehicle and the target object according to the risk level corresponding to the sub-risk area comprises:
and determining the danger level corresponding to the sub-danger area according to the width of the sub-danger area.
5. The method of claim 1, wherein determining the location of the bounding box corresponding to the target object according to the type, length, and width of the target object comprises:
determining the reference point position of the bounding box corresponding to the target object according to the type, the length and the width of the target object;
and determining the position of the bounding box corresponding to the target object according to the position of the reference point.
6. The method of claim 5, wherein the types of the target objects comprise motor vehicles, non-motor vehicles and pedestrians, and the determining the reference point position of the bounding box corresponding to the target object according to the type, length and width of the target object comprises:
when the type of the target object is a motor vehicle, determining the center of a head line or the center of a tail line of the motor vehicle as the reference point position of the bounding box corresponding to the target object;
when the type of the target object is a non-motor vehicle or a pedestrian, determining the geometric center of the non-motor vehicle or the pedestrian as the reference point position of the bounding box corresponding to the target object.
7. A collision risk prediction apparatus, characterized by comprising:
the acquisition module is used for acquiring attribute information and motion attitude information of the target object; the target object is an object whose distance from the current vehicle is less than or equal to a first threshold value; the attribute information comprises the type, length and width of the target object;
the surrounding module is used for determining the size of a target surrounding box corresponding to the target object according to the width of the target object; determining the number of the target bounding boxes corresponding to the target object according to the length of the target object; determining the position of the target bounding box corresponding to the target object according to the type, the length and the width of the target object; the target bounding box covers the target object;
the first determining module is used for determining the motion track of the target bounding box according to the motion attitude information of the target object to obtain a first motion track;
the second determining module is used for determining the motion track of the bounding box of the current vehicle according to the motion attitude information of the current vehicle and the attribute information of the current vehicle to obtain a second motion track;
and the output module is used for determining the collision probability of the current vehicle and the target object according to the first motion trail and the second motion trail.
8. A computer device, characterized in that the computer device comprises: a processor and a memory; the memory for storing computer program code, the computer program code comprising computer instructions; the computer device, when executing the computer instructions, performs a collision risk prediction method according to any of claims 1-6.
9. A computer-readable storage medium comprising computer instructions that, when executed on a computer device, cause the computer device to perform the collision risk prediction method of any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111593273.8A CN113978457B (en) | 2021-12-24 | 2021-12-24 | Collision risk prediction method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111593273.8A CN113978457B (en) | 2021-12-24 | 2021-12-24 | Collision risk prediction method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113978457A CN113978457A (en) | 2022-01-28 |
CN113978457B true CN113978457B (en) | 2022-05-03 |
Family
ID=79734161
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111593273.8A Active CN113978457B (en) | 2021-12-24 | 2021-12-24 | Collision risk prediction method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113978457B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114913626A (en) * | 2022-05-07 | 2022-08-16 | 中汽创智科技有限公司 | Data processing method, device, equipment and storage medium |
CN115083208B (en) * | 2022-07-20 | 2023-02-03 | 深圳市城市交通规划设计研究中心股份有限公司 | Human-vehicle conflict early warning method, early warning analysis method, electronic device and storage medium |
CN115472005B (en) * | 2022-08-09 | 2023-12-19 | 东软睿驰汽车技术(上海)有限公司 | Vehicle collision early warning method, device, equipment and storage medium |
CN115782867B (en) * | 2022-11-17 | 2024-01-30 | 上海西井科技股份有限公司 | Track collision risk assessment method, device, electronic equipment and storage medium |
CN117657133A (en) * | 2023-02-20 | 2024-03-08 | 比亚迪股份有限公司 | Vehicle safety control method, device and system and vehicle |
CN115837921B (en) * | 2023-02-23 | 2023-06-02 | 深圳佑驾创新科技有限公司 | Vehicle track collision detection method, device, equipment and storage medium |
CN117114547B (en) * | 2023-10-25 | 2024-02-02 | 交通运输部水运科学研究所 | Harbor enterprise cluster type safety responsibility early warning method and system based on cloud platform |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102275587A (en) * | 2011-06-07 | 2011-12-14 | 长安大学 | Rear vehicle collision danger monitoring device and monitoring method thereof |
CN111731283A (en) * | 2020-05-26 | 2020-10-02 | 北京百度网讯科技有限公司 | Vehicle collision risk identification method and device and electronic equipment |
CN112026762A (en) * | 2020-09-15 | 2020-12-04 | 北京经纬恒润科技有限公司 | Vehicle collision analysis method and device |
WO2021034802A1 (en) * | 2019-08-22 | 2021-02-25 | Argo AI, LLC | Systems and methods for trajectory based safekeeping of vehicles |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8466807B2 (en) * | 2011-06-01 | 2013-06-18 | GM Global Technology Operations LLC | Fast collision detection technique for connected autonomous and manual vehicles |
DE102013010004A1 (en) * | 2013-06-14 | 2014-12-18 | Valeo Schalter Und Sensoren Gmbh | Method and device for carrying out collision avoiding measures |
US9280899B2 (en) * | 2013-08-06 | 2016-03-08 | GM Global Technology Operations LLC | Dynamic safety shields for situation assessment and decision making in collision avoidance tasks |
DE102015205048A1 (en) * | 2015-03-20 | 2016-09-22 | Robert Bosch Gmbh | Method and device for monitoring a target trajectory to be traveled by a vehicle for collision freedom |
US10077047B2 (en) * | 2017-02-10 | 2018-09-18 | Waymo Llc | Using wheel orientation to determine future heading |
WO2019136479A1 (en) * | 2018-01-08 | 2019-07-11 | The Regents On The University Of California | Surround vehicle tracking and motion prediction |
US11079764B2 (en) * | 2018-02-02 | 2021-08-03 | Nvidia Corporation | Safety procedure analysis for obstacle avoidance in autonomous vehicles |
CN111133448A (en) * | 2018-02-09 | 2020-05-08 | 辉达公司 | Controlling autonomous vehicles using safe arrival times |
CN108725440B (en) * | 2018-04-20 | 2020-11-27 | 深圳市商汤科技有限公司 | Forward collision control method and apparatus, electronic device, program, and medium |
US11001256B2 (en) * | 2018-09-19 | 2021-05-11 | Zoox, Inc. | Collision prediction and avoidance for vehicles |
US11167754B2 (en) * | 2019-08-22 | 2021-11-09 | Argo AI, LLC | Systems and methods for trajectory based safekeeping of vehicles |
US20210197805A1 (en) * | 2019-12-27 | 2021-07-01 | Motional Ad Llc | Safety system for vehicle |
EP3886076A1 (en) * | 2020-03-27 | 2021-09-29 | Aptiv Technologies Limited | Warning system for a host automotive vehicle |
CN111507222B (en) * | 2020-04-09 | 2023-07-07 | 中山大学 | Three-dimensional object detection frame based on multisource data knowledge migration |
US20210339741A1 (en) * | 2020-04-30 | 2021-11-04 | Zoox, Inc. | Constraining vehicle operation based on uncertainty in perception and/or prediction |
CN111707269B (en) * | 2020-06-23 | 2022-04-05 | 东南大学 | Unmanned aerial vehicle path planning method in three-dimensional environment |
-
2021
- 2021-12-24 CN CN202111593273.8A patent/CN113978457B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102275587A (en) * | 2011-06-07 | 2011-12-14 | 长安大学 | Rear vehicle collision danger monitoring device and monitoring method thereof |
WO2021034802A1 (en) * | 2019-08-22 | 2021-02-25 | Argo AI, LLC | Systems and methods for trajectory based safekeeping of vehicles |
CN111731283A (en) * | 2020-05-26 | 2020-10-02 | 北京百度网讯科技有限公司 | Vehicle collision risk identification method and device and electronic equipment |
CN112026762A (en) * | 2020-09-15 | 2020-12-04 | 北京经纬恒润科技有限公司 | Vehicle collision analysis method and device |
Non-Patent Citations (1)
Title |
---|
人车碰撞风险识别及智能车辆控制***;孟柯等;《交通信息与安全》;20161228(第06期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113978457A (en) | 2022-01-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113978457B (en) | Collision risk prediction method and device | |
US10488863B2 (en) | Autonomous vehicle post-fault operation | |
CN107918758B (en) | Vehicle capable of environmental scenario analysis | |
CN111309013B (en) | Collision distance determining method and system, vehicle and storage medium | |
JP4416020B2 (en) | Travel plan generator | |
CN108919802B (en) | Unmanned vehicle driving method and device | |
US10829113B2 (en) | Vehicle collision avoidance | |
EP1227302B1 (en) | Bowing coefficient representation of curvature of geographic features | |
KR102042111B1 (en) | Driving assistance apparatus | |
US11845471B2 (en) | Travel assistance method and travel assistance device | |
KR102086270B1 (en) | Control method and traveling control device of the traveling control device | |
CN111179624A (en) | Collision early warning method and device, computer equipment and storage medium | |
EP3831679A1 (en) | Traveling trajectory correction method, traveling control method, and traveling trajectory correction device | |
EP3456596A1 (en) | Method and device of predicting a possible collision | |
US11238731B2 (en) | Method and device for detecting the passage of a motor vehicle through a road sign gantry | |
US11042160B2 (en) | Autonomous driving trajectory determination device | |
KR20190045308A (en) | A vehicle judging method, a traveling path correcting method, a vehicle judging device, and a traveling path correcting device | |
US10095238B2 (en) | Autonomous vehicle object detection | |
EP3925845A1 (en) | Other vehicle action prediction method and other vehicle action prediction device | |
CN118235180A (en) | Method and device for predicting drivable lane | |
CN114291116B (en) | Surrounding vehicle track prediction method and device, vehicle and storage medium | |
JP2017154516A (en) | Collision determination device, collision determination method, and program | |
JP7259818B2 (en) | self-driving device | |
KR20180006635A (en) | System for avoiding collision in crossroad and method for control thereof | |
CN116161111B (en) | Vehicle control method and device, vehicle and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP03 | Change of name, title or address |
Address after: Floor 25, Block A, Zhongzhou Binhai Commercial Center Phase II, No. 9285, Binhe Boulevard, Shangsha Community, Shatou Street, Futian District, Shenzhen, Guangdong 518000 Patentee after: Shenzhen Youjia Innovation Technology Co.,Ltd. Address before: 518051 401, building 1, Shenzhen new generation industrial park, No. 136, Zhongkang Road, Meidu community, Meilin street, Futian District, Shenzhen, Guangdong Province Patentee before: SHENZHEN MINIEYE INNOVATION TECHNOLOGY Co.,Ltd. |
|
CP03 | Change of name, title or address |