CN115214724B - Trajectory prediction method and apparatus, electronic device and storage medium - Google Patents
Trajectory prediction method and apparatus, electronic device and storage medium Download PDFInfo
- Publication number
- CN115214724B CN115214724B CN202211140393.7A CN202211140393A CN115214724B CN 115214724 B CN115214724 B CN 115214724B CN 202211140393 A CN202211140393 A CN 202211140393A CN 115214724 B CN115214724 B CN 115214724B
- Authority
- CN
- China
- Prior art keywords
- obstacle
- vehicle
- determining
- self
- virtual
- 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 50
- 230000003993 interaction Effects 0.000 claims abstract description 98
- 230000002452 interceptive effect Effects 0.000 claims abstract description 54
- 230000004888 barrier function Effects 0.000 claims description 58
- 238000004891 communication Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000011144 upstream manufacturing Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 241000283070 Equus zebra Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004044 response Effects 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
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0027—Planning or execution of driving tasks using trajectory prediction for other traffic participants
-
- 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
- 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
- B60W50/0097—Predicting future 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
- 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
- B60W50/0098—Details of control systems ensuring comfort, safety or stability not otherwise provided for
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/056—Detecting movement of traffic to be counted or controlled with provision for distinguishing direction of travel
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention relates to the technical field of trajectory prediction, and discloses a trajectory prediction method, a trajectory prediction device, electronic equipment and a storage medium, wherein the method comprises the following steps: determining the speed direction of the obstacle and the speed direction of the vehicle according to the acquired obstacle motion parameters and the acquired vehicle motion parameters; determining the type of the obstacle to which the obstacle belongs according to the relation between the included angle of the two and a preset angle range; when the type of the obstacle is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle; under the condition that an interaction intention exists between the obstacle and the self-vehicle, determining a virtual prediction track which belongs to the obstacle and is over against the self-vehicle through an interaction motion model, an obstacle motion parameter and a self-vehicle motion parameter; and under the condition that no interaction intention exists between the obstacle and the self vehicle, determining the virtual predicted track of the obstacle through the non-interactive motion model and the obstacle motion parameters. The aim is to avoid frequent inching and braking of the vehicle so as to improve the running stability and the running efficiency of the vehicle.
Description
Technical Field
The present invention relates to the field of trajectory prediction technologies, and in particular, to a method and an apparatus for trajectory prediction, an electronic device, and a storage medium.
Background
Vehicle interaction with pedestrians and bicycles is unavoidable. The main operational scenario of a passenger vehicle is a motorway, which is relatively regular in interaction with pedestrians and bicycles. Most pedestrians and bicycles can interact with the bicycle when crossing the zebra crossing, and the irregular road-crossing behavior is relatively small in quantity. The main operation scene of the unmanned logistics vehicle for sending the express is a side road, the running speed of the unmanned logistics vehicle is slow (about 20 km/h), and the obstacles which interact with the logistics vehicle most on the side road are pedestrians and bicycles. The pedestrian and the bicycle have the characteristics of higher degree of freedom, high collision cost and the like. If the two types of obstacles cannot be predicted with high quality, the vehicle can be frequently braked, the operation efficiency is influenced due to the fact that the vehicle is heavy, and the risk of safety accidents is caused.
In order to solve the above problems, it is common practice to perform kinematic modeling on pedestrian and bicycle obstacles, such as using a uniform velocity linear motion model or a uniform acceleration linear motion model. Or generating a track by using a data-driven model and capturing motion characteristics by using the complexity of a network to generate a more complex track. Although the realization of a common uniform-speed linear motion or uniform-acceleration linear motion model is simple, the effect is controllable, the stopping or steering behavior in the middle of the barrier cannot be captured, and meanwhile, the uniform-speed linear motion model has poor tolerance on the sensed speed error and fluctuates along with the sensed fluctuation, so that a plurality of unnecessary braking, namely abnormal point braking, can be caused. Although the data driving model is complex and has excellent indexes on a large data set, the problem can not be solved and the safety can not be ensured. Meanwhile, tracks generated by the two models accord with an objective motion rule, but a short plate of an s-t graph algorithm of a downstream planning module is not considered, namely the s-t graph algorithm is only sensitive to longitudinal tracks and has almost no response to transverse tracks vertical to the longitudinal direction of the self-vehicle. Even if the future movement condition of the crosscut obstacle is well predicted by using the two methods, the self vehicle collides with the obstacle or generates a collision risk because the planning module is insensitive to the crosscut track.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus, an electronic device and a storage medium for trajectory prediction. Aims to avoid frequent inching and braking of the vehicle so as to improve the running stability and the running efficiency of the vehicle.
The invention provides a track prediction method, which comprises the following steps:
determining the speed direction of the obstacle and the speed direction of the self-vehicle according to the acquired obstacle motion parameters and the self-vehicle motion parameters;
determining the type of the obstacle to which the obstacle belongs according to the relation between the included angle between the speed direction of the obstacle and the speed direction of the vehicle and a preset angle range, wherein the preset angle range comprises a first preset angle range, a second preset angle range and a third preset angle range;
when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle;
under the condition that the barrier and the self vehicle have interactive intention, determining a virtual predicted track which belongs to the barrier and is over against the self vehicle through an interactive motion model, the barrier motion parameter and the self vehicle motion parameter;
and under the condition that no interaction intention exists between the obstacle and the self vehicle, determining a virtual predicted track of the obstacle through a non-interactive motion model and the obstacle motion parameters.
Optionally, when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether there is an interaction intention between the obstacle and the host vehicle includes:
when the type of the obstacle belongs to the opposite obstacle, determining a first included angle between the speed direction of the obstacle and a transverse axis, and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis;
determining that an interaction intention exists between the barrier and the self-vehicle under the condition that the first included angle is smaller than or equal to the second included angle;
and under the condition that the first included angle is larger than the second included angle, determining that no interaction intention exists between the barrier and the self-vehicle.
Optionally, the determining a virtual predicted trajectory of the obstacle through a non-interactive motion model and the obstacle motion parameter in the case that there is no interactive intention between the obstacle and the own vehicle includes:
determining the component speed of the obstacle opposite to the moving direction of the self-vehicle through the obstacle moving parameters;
and constructing a virtual predicted track of the obstacle, wherein the longitudinal coordinate of the virtual predicted track is the same as that of the position of the obstacle at the current moment, and the starting point of the virtual predicted track is located at the position of the obstacle at the current moment.
Optionally, when the type of the obstacle to which the obstacle belongs is a crosscut obstacle, determining the time length required for the obstacle to cut into the front of the vehicle according to the obstacle motion parameter;
when the duration meets a first preset condition, determining a virtual predicted track which belongs to the barrier and is over against the self-vehicle through an interactive motion model, the barrier motion parameter and the self-vehicle motion parameter;
and when the duration does not meet a first preset condition, determining the predicted track of the obstacle through a uniform linear motion model and the obstacle motion parameter.
Optionally, when the type of the obstacle to which the obstacle belongs is an opposite obstacle, before determining whether there is an interaction intention between the obstacle and the host vehicle, the method further includes:
taking the position of the self-vehicle as an original point, and carrying out rasterization division on the periphery of the self-vehicle by preset sizes to obtain a plurality of grid areas;
determining the grid region types to which the grid regions belong according to the position relations between the grid regions and the own vehicle;
when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle comprises the following steps:
when the type of the obstacle belongs to the opposite obstacle, determining the grid area type of the obstacle; determining that the obstacle has an interaction intention with the host vehicle if the obstacle is within a grid area of a first grid area category; under the condition that the obstacle is in a grid area of a second grid area type, determining a first included angle between the speed direction of the obstacle and a transverse axis, and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis; determining that an interaction intention exists between the barrier and the vehicle under the condition that the first included angle is smaller than or equal to the second included angle, or determining that the interaction intention does not exist between the barrier and the vehicle under the condition that the first included angle is larger than the second included angle; determining that the obstacle has a target interaction intention with the host vehicle if the obstacle is within a grid area of a third grid area category;
and under the condition that the obstacle and the self-vehicle have an interaction intention, determining the predicted track of the obstacle through a uniform-speed linear motion model and the obstacle motion parameters so as to sensitively respond to the motion direction of the obstacle.
Optionally, the determining, in a case where there is an interaction intention between the obstacle and the own vehicle, a virtual predicted trajectory belonging to the obstacle and facing the own vehicle through an interaction motion model, the obstacle motion parameter, and the own vehicle motion parameter includes:
determining the partial speed of the obstacle opposite to the moving direction of the self vehicle through the obstacle moving parameter under the condition that the obstacle and the self vehicle have the interaction intention;
determining a virtual intersection point of the self vehicle and the obstacle on a longitudinal axis of the self vehicle according to the partial speed, the speed of the self vehicle at the current moment and the distance between the obstacle and the self vehicle in the longitudinal direction;
and constructing a virtual prediction track with the longitudinal coordinate same as the longitudinal coordinate of the self vehicle and the track end point as the virtual intersection point through the partial speed and the virtual intersection point.
Optionally, when the duration satisfies a first preset condition, determining, through an interactive motion model, the obstacle motion parameter, and the vehicle motion parameter, a virtual predicted trajectory belonging to the obstacle and facing the vehicle includes:
when the duration meets a first preset condition, determining a virtual intersection point where the barrier and a longitudinal axis where the self vehicle is located intersect through the speed direction of the barrier;
constructing a virtual prediction track with the longitudinal coordinate same as the longitudinal coordinate of the self-vehicle and the track end point as the virtual intersection point through the preset speed of the barrier and the virtual intersection point
Aiming at the prior art, the invention has the following advantages:
according to the track prediction method provided by the invention, the speed direction of the obstacle and the speed direction of the self-vehicle are determined according to the relation between the included angle between the speed direction of the obstacle and the speed direction of the self-vehicle and a preset angle range according to the acquired obstacle motion parameters and the self-vehicle motion parameters, and the type of the obstacle to which the obstacle belongs is determined; when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle; under the condition that the barrier and the self-vehicle have interactive intention, a virtual prediction track which belongs to the barrier and is right opposite to the self-vehicle is constructed through an interactive motion model, the barrier motion parameters and the self-vehicle motion parameters, so that a downstream decision end can timely make a deceleration obstacle avoidance action based on the virtual prediction track right opposite to the self-vehicle, the self-vehicle is prevented from frequently stopping, and the running stability and the running efficiency of the self-vehicle are improved. And meanwhile, under the condition that no interaction intention exists between the barrier and the self vehicle, a virtual prediction track parallel to the running track of the self vehicle and belonging to the barrier is determined through a non-interactive motion model and the motion parameters of the barrier, so that the decision of the downstream decision end of the self vehicle, which is contrary to the decision of the upstream track prediction end, is avoided.
The second aspect of the present invention provides a trajectory prediction apparatus. Aims to avoid frequent inching and braking of the vehicle so as to improve the running stability and the running efficiency of the vehicle.
The invention provides a track prediction device, which comprises:
the speed direction determining unit is used for determining the speed direction of the obstacle and the speed direction of the self-vehicle according to the acquired obstacle motion parameters and the self-vehicle motion parameters;
the obstacle type identification unit is used for determining the type of the obstacle to which the obstacle belongs according to the relation between the included angle between the speed direction of the obstacle and the speed direction of the vehicle and a preset angle range, wherein the preset angle range comprises a first preset angle range, a second preset angle range and a third preset angle range;
an interaction intention determining unit, configured to determine whether there is an interaction intention between the obstacle and the host vehicle when the obstacle type to which the obstacle belongs is an opposing obstacle;
a first virtual predicted track generation unit, configured to determine, through an interactive motion model, the obstacle motion parameter, and the vehicle motion parameter, a virtual predicted track that belongs to the obstacle and faces the vehicle when there is an interactive intention between the obstacle and the vehicle;
and the second virtual predicted track generating unit is used for determining the virtual predicted track of the obstacle through a non-interactive motion model and the obstacle motion parameters under the condition that no interactive intention exists between the obstacle and the self vehicle.
The third aspect of the present invention provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus;
a memory for storing a computer program;
a processor, configured to implement the steps of the method for trajectory prediction according to the first aspect when executing the program stored in the memory.
A fourth aspect of the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps in a method of trajectory prediction as described in the first aspect above.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of a method for trajectory prediction according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a position relationship between an obstacle and a host vehicle in a trajectory prediction method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a predetermined angle range in a trajectory prediction method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating interactive intent determination in a trajectory prediction method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a virtual predicted trajectory with an interaction intention in a trajectory prediction method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a virtual predicted trajectory without an interaction intention in a trajectory prediction method according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a virtual predicted trajectory of a lateral obstacle in a trajectory prediction method according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of rasterization in a trajectory prediction method provided by an embodiment of the present invention;
fig. 9 is a schematic diagram of an apparatus for trajectory prediction according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
Before the description of the invention, the application scene of the invention is described as an unmanned logistics vehicle for sending express, the main operation scene is a side road, the driving speed is slow, about 20km/h, and the obstacles on the side road most interacting with the logistics vehicle are pedestrians and bicycles.
Fig. 1 is a flowchart of a method for trajectory prediction according to an embodiment of the present invention, and as shown in fig. 1, the method includes:
step S101: determining the speed direction of the obstacle and the speed direction of the self-vehicle according to the acquired obstacle motion parameters and the self-vehicle motion parameters;
step S102: determining the type of the obstacle to which the obstacle belongs according to the relation between the included angle between the speed direction of the obstacle and the speed direction of the vehicle and a preset angle range, wherein the preset angle range comprises a first preset angle range, a second preset angle range and a third preset angle range;
step S103: when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle;
step S104: under the condition that the barrier and the self vehicle have interactive intention, determining a virtual predicted track which belongs to the barrier and is over against the self vehicle through an interactive motion model, the barrier motion parameter and the self vehicle motion parameter;
step S105: and under the condition that no interaction intention exists between the obstacle and the self vehicle, determining a virtual predicted track of the obstacle through a non-interactive motion model and the obstacle motion parameters.
In this embodiment, the motion parameters of the obstacle are acquired by a sensor of the vehicle, and the acquired motion parameters of the obstacle are analyzed to obtain the position relationship of the obstacle with respect to the vehicle, the speed and the speed direction of the obstacle, and the like. The speed, the speed direction and the like of the vehicle are obtained by analyzing and processing the motion parameters of the vehicle. The coordinate system describing the position relationship between the vehicle and the obstacle is a Frenet coordinate system, the coordinate system takes the middle position of the head of the vehicle as a coordinate origin, the motion direction of the vehicle as a longitudinal axis S, and the left side perpendicular to the motion direction of the vehicle as a horizontal axis L.
As shown in fig. 2, EGO is the vehicle, v1 is the speed of the vehicle, and the coordinate system is the vertical axis S axis, so the direction of the speed of the vehicle will always be in the same direction as the vertical axis S axis, a circle is an obstacle, and v2 is the speed of the obstacle.
Three preset angle ranges including a first preset angle range, a second preset angle range and a third preset angle range are set in advance, and the three preset angle ranges are used for determining the type of the obstacle to which the obstacle belongs.
When the included angle between the speed direction of the obstacle and the speed direction of the vehicle is within a first preset angle range, determining the obstacle as an opposite obstacle; when the included angle between the speed direction of the obstacle and the speed direction of the vehicle is within a second preset angle range, determining the obstacle as a transverse obstacle; and when the included angle between the speed direction of the obstacle and the speed direction of the self-vehicle is within a third preset angle range, determining that the obstacle is a homodromous obstacle.
In this embodiment, the values of the first preset angle range are preferably [110 °,180 ° ], the values of the second preset angle range are preferably [80 °,110 °) and the values of the third preset angle range are preferably [0 °,80 °), it should be understood that the values of the first preset angle range, the second preset angle range and the third preset angle range are only a preferred embodiment, and the values of the three preset angle ranges may also be other ranges, and are not limited specifically herein.
As shown in fig. 3, the range formed by the included angles C11 and C12 is the third preset angle range, the range formed by the included angles C21 and C22 is the second preset angle range, and the range formed by the included angles C31 and C32 is the first preset angle range. When the first, second and third preset angle ranges are the above preferred embodiment, the included angles C11 and C12 in fig. 3 are [0 ° and 80 °), the included angles C21 and C22 in fig. 3 are [80 ° and 110 °), and the included angles C31 and C32 in fig. 3 are [110 ° and 180 ° ], respectively.
When it is determined that the type of the obstacle to which the obstacle belongs is the opposing obstacle, it is determined whether there is an intention of interaction between the obstacle and the own vehicle, that is, whether there is a possibility that the obstacle is present in collision with the own vehicle.
In the present invention, when the type of the obstacle to which the obstacle belongs is an opposing obstacle, determining whether there is an interaction intention between the obstacle and the host vehicle includes: when the type of the obstacle belongs to the opposite obstacle, determining a first included angle between the speed direction of the obstacle and a transverse axis, and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis; determining that an interaction intention exists between the barrier and the self-vehicle under the condition that the first included angle is smaller than or equal to the second included angle; and under the condition that the first included angle is larger than the second included angle, determining that no interaction intention exists between the barrier and the self-vehicle.
In this embodiment, when the type of the obstacle to which the obstacle belongs is an opposing obstacle, a specific implementation manner of determining whether there is an interaction intention between the obstacle and the host vehicle is to determine an angle between a speed direction of the obstacle and a horizontal axis, which is referred to as a first angle, and determine an angle between a line connecting the obstacle and the host vehicle and the horizontal axis, which is referred to as a second angle.
And determining that the interaction intention exists between the barrier and the self-vehicle under the condition that the first included angle is smaller than or equal to the second included angle, and determining that the interaction intention does not exist between the barrier and the self-vehicle under the condition that the first included angle is larger than the second included angle.
For example, as shown in fig. 4, when the obstacle A1 travels in the speed direction of V21, the first angle between the speed direction of the obstacle A1 and the horizontal axis is determined to be C1, the second angle between the horizontal axis and the line connecting the obstacle A1 and the host vehicle is determined to be C2, and C1 is smaller than C2, so that when the obstacle A1 travels in the speed direction of V21, there is a possibility of collision with the host vehicle, and there is an interaction intention between the obstacle and the host vehicle.
When the obstacle A1 travels in the direction of the speed V22, the first included angle between the speed direction of the obstacle A1 and the horizontal axis is determined to be C3, the second included angle between the horizontal axis and the line connecting the obstacle A1 and the host vehicle is determined to be C2, and at this time, C3 is greater than C2, so that when the obstacle A1 travels in the direction of the speed V22, there is no possibility of collision with the host vehicle, and there is no interaction intention between the obstacle and the host vehicle.
When the obstacle A2 travels in the speed direction of V31, the first angle between the speed direction of the obstacle A2 and the horizontal axis is determined to be C5, the second angle between the line connecting the obstacle A2 and the host vehicle and the horizontal axis is determined to be C4, and at this time, C5 is smaller than C4, so that when the obstacle A2 travels in the speed direction of V31, there is a possibility of collision with the host vehicle, and there is an interaction intention between the obstacle and the host vehicle.
When the obstacle A2 moves towards the direction of the speed V32, the first included angle between the speed direction of the obstacle A2 and the horizontal axis is determined to be C6, the second included angle between the horizontal axis and the line connecting the obstacle A2 and the host vehicle is determined to be C4, and at this time, C6 is greater than C4, so that when the obstacle A2 moves towards the direction of the speed V32, the possibility of collision with the host vehicle does not exist, and the interaction intention does not exist between the obstacle and the host vehicle.
In the embodiment, under the condition that the interaction intention between the obstacle and the vehicle is determined, the virtual predicted track of the obstacle is determined through the interaction motion model and the obstacle motion parameter.
In the present invention, the determining, by an interactive motion model, the obstacle motion parameter, and the vehicle motion parameter, a virtual predicted trajectory belonging to the obstacle and facing the vehicle includes: determining the partial speed of the obstacle opposite to the moving direction of the self vehicle through the obstacle moving parameter under the condition that the obstacle and the self vehicle have the interaction intention; determining a virtual intersection point of the self vehicle and the obstacle on a longitudinal axis of the self vehicle according to the partial speed, the speed of the self vehicle at the current moment and the distance between the obstacle and the self vehicle in the longitudinal direction; and constructing a virtual prediction track with the longitudinal coordinate same as the longitudinal coordinate of the self vehicle and the track end point as the virtual intersection point through the partial speed and the virtual intersection point.
In this embodiment, a specific implementation manner of determining the virtual predicted track, which belongs to the obstacle and is directly facing the self-vehicle, through the interactive motion model and the obstacle motion parameter is to construct a virtual predicted track for the obstacle based on the obstacle motion parameter and the self-vehicle motion parameter by calling the interactive motion model, and a specific implementation process thereof is as follows.
And determining the speed direction and the speed of the obstacle at the current moment through the obstacle parameters, and determining the component speed of the obstacle opposite to the moving direction of the vehicle according to the speed direction and the speed of the obstacle. After obtaining the partial velocity of the obstacle, a virtual intersection point of the vehicle and the obstacle on the longitudinal axis S of the vehicle is determined by the partial velocity of the obstacle, the velocity of the vehicle at the present time, the distance between the obstacle and the vehicle in the longitudinal direction, and the following kinematic formula (1).
Wherein v is ego The speed of the current time of the self-vehicle belongs to a known quantity; v. of s The speed component of the obstacle in the direction opposite to the moving direction of the self-vehicle belongs to a known quantity; s dis-S The distance between the obstacle and the self-vehicle in the longitudinal direction belongs to a known quantity; a is the deceleration of the vehicle and belongs to an unknown quantity; t is a time period required for the host vehicle to travel at the above deceleration and finally stop at the virtual intersection, and is an unknown quantity.
Calculated by the above two calculation formulasBased on the values of the deceleration a and the time length t of the vehicle, the distance S = v at which the virtual intersection is located directly in front of the vehicle is obtained ego X t +1/2at2, thereby obtaining a virtual intersection.
And after the virtual intersection point is obtained, constructing a virtual predicted track of the obstacle with the longitudinal coordinate same as the longitudinal coordinate of the vehicle by taking the partial speed of the obstacle and the virtual intersection point as the end point of the virtual predicted track.
For example, as shown in fig. 5, the obstacle motion parameter is analyzed to obtain a velocity V52 of the obstacle having a velocity direction at the current time, and a component velocity V52_1 of the obstacle in the direction opposite to the moving direction of the vehicle is obtained by calculating the velocity V52 of the obstacle. The speed of the vehicle at the current moment and the distance S between the obstacle and the vehicle in the longitudinal direction are obtained by the partial speed of the obstacle dis-S And the kinematic formula (1) determines a virtual intersection point P1 of the vehicle and the obstacle on the longitudinal axis S of the vehicle. The method includes the steps that a virtual predicted track of an obstacle with the vertical coordinate same as the vertical coordinate of the position of a self-vehicle is constructed by the partial speed V52_1 of the obstacle and the virtual intersection point P1, meanwhile, the track end point is the virtual intersection point P1, the obstacle is preset to always run along the virtual predicted track by the partial speed V52_1 in the process, therefore, the starting point of the virtual predicted track is the horizontal coordinate same as the horizontal coordinate of the position of the obstacle at the current moment, the vertical coordinate is the point P2 same as the vertical coordinate of the position of the self-vehicle, and as shown in fig. 5, the virtual predicted track is a line segment P2-P1 in fig. 5.
It should be understood that the virtual predicted trajectory of the obstacle generated by the interactive motion model mentioned above is a virtual predicted trajectory that is not associated with any actual trajectory of the obstacle, and is only related to the speed orientation, speed and position coordinates of the obstacle at the current time. After the virtual predicted track of the obstacle is sent to the downstream decision end by the upstream, the downstream considers that the motion track of the obstacle is the virtual predicted track of the obstacle, and a corresponding control strategy is made based on the virtual predicted track without paying attention to the actual motion track of the obstacle.
In this embodiment, the previous uniform velocity linear motion model or uniform acceleration linear motion model has a poor tolerance to the sensed velocity error, and fluctuates along with the fluctuation of the sensing, so that many unnecessary braking stops are caused, that is, as long as the motion parameter of the obstacle sensed by the vehicle fluctuates, the vehicle is easily switched between braking and non-braking based on the sensed motion parameter of the obstacle, and the predicted trajectory generated by the uniform velocity linear motion model or uniform acceleration linear motion model causes frequent braking, thereby reducing the stability of the vehicle and the operating efficiency. The virtual predicted track determined by the embodiment is generated when the intention of interaction between the obstacle and the self-vehicle exists, and the judgment of the interaction intention is determined through an angle range, so that the virtual predicted track for controlling the self-vehicle to make a deceleration obstacle avoidance decision can be generated as long as the speed and the speed direction of the obstacle sensed have certain errors within a reasonable range of the interaction intention, the self-vehicle can not make frequent spot braking decisions like the prior art, the motion stability of the self-vehicle is higher, and the running efficiency is improved. Or in a reasonable range without the interaction intention, a virtual prediction track for controlling the self-vehicle to make a deceleration obstacle avoidance decision cannot be generated, so that the motion stability of the self-vehicle is higher, and the running efficiency is higher.
In the embodiment, under the condition that the fact that no interaction intention exists between the obstacle and the own vehicle is determined, the virtual predicted track of the obstacle is determined through the non-interactive motion model and the obstacle motion parameters.
In the present invention, the determining a virtual predicted trajectory of the obstacle through a non-interactive motion model and the obstacle motion parameter in the case that there is no interactive intention between the obstacle and the own vehicle includes: determining the component speed of the obstacle opposite to the movement direction of the self vehicle through the obstacle movement parameters; and constructing a virtual prediction track of the obstacle, wherein the longitudinal coordinate of the virtual prediction track is the same as that of the position of the obstacle at the current moment, and the starting point of the virtual prediction track is located at the position of the obstacle at the current moment.
In the embodiment, for the obstacle determined as the type of the opposite-direction obstacle, although it is determined that there is no interaction intention between the obstacle and the host vehicle by the above-described embodiment, if there is a trajectory segment that is relatively close to the host vehicle in the predicted trajectory of the obstacle generated directly by the uniform linear motion model, the obstacle avoidance may be considered in the downstream decision making process, while it is actually determined that there is no interaction intention between the obstacle and the host vehicle in the above-described embodiment, so that the decision making process is contrary to the decision making process in the downstream. That is, the upstream already determines that the obstacle is an obstacle which will not collide with the own vehicle, and the downstream makes an obstacle avoidance decision when receiving a predicted track of the obstacle generated by the upstream through a uniform linear motion model. Therefore, to avoid this situation, the present invention generates a virtual predicted trajectory of the obstacle by using a non-interactive motion model and obstacle motion parameters, specifically:
and analyzing the obstacle motion parameters to obtain the speed and the speed direction of the obstacle, and calculating the speed and the speed direction of the obstacle to obtain the component speed of the obstacle in the direction opposite to the moving direction of the vehicle. And constructing a virtual prediction track with the longitudinal coordinate same as that of the position of the obstacle at the current moment from the current position of the obstacle based on the component speed of the obstacle.
For example, as shown in fig. 6, the obstacle motion parameter is analyzed to obtain the speed and the speed direction of the obstacle, and the speed direction of the obstacle are calculated to obtain the partial speed V62_1 of the obstacle in the direction opposite to the moving direction of the vehicle. Based on the partial velocity V62_1 of the obstacle, starting from the current position of the obstacle, a virtual predicted trajectory is constructed, such as L2 in fig. 6, whose longitudinal coordinate is the same as the longitudinal coordinate of the current position of the obstacle.
By means of the implementation mode, when it is determined that no interaction intention exists between an obstacle belonging to the type of the opposite obstacle and the vehicle at the upstream, the virtual predicted track of the obstacle at the downstream is given by the upstream, and the track of the obstacle at the downstream is considered to be always kept at the same distance from the vehicle at the downstream regardless of the actual motion track of the obstacle, so that obstacle avoidance decisions cannot be made on the obstacle, and the decision that the obstacle is violated at the downstream is avoided.
In the invention, when the type of the obstacle belongs to the crosscut obstacle, the time length required for the obstacle to cut into the front of the bicycle is determined according to the movement parameters of the obstacle; when the duration meets a first preset condition, determining a virtual predicted track which belongs to the barrier and is over against the self-vehicle through an interactive motion model and the barrier motion parameters; and when the duration does not meet a first preset condition, determining the predicted track of the obstacle through a uniform linear motion model and the obstacle motion parameter.
In the present embodiment, when it is determined that the obstacle belongs to a cross obstacle, that is, an obstacle cut into the front of the vehicle from the front of the vehicle in the lateral direction. For such an obstacle, only an obstacle in which the vehicle is cut into right in front of the vehicle within a set period of time needs to be paid attention to. Specifically, the barrier motion parameters are analyzed to obtain the velocity and the velocity direction of the barrier, and the velocity direction of the barrier are calculated to obtain the component velocity of the barrier along the horizontal axis direction. And determining the time length required by the obstacle to cut into the front of the vehicle according to the component speed of the obstacle along the horizontal axis direction and the horizontal axis coordinate of the obstacle.
And when the determined duration is higher than the set duration, determining that the duration required by the obstacle to cut into the front of the vehicle does not meet the first set condition, and the vehicle can temporarily pay no attention to the obstacle, wherein the predicted track of the obstacle can be constructed through the uniform-speed linear model only by the speed and the speed direction of the obstacle.
And when the determined duration is less than the set duration, determining that the duration required by the obstacle to cut into the front of the self meets a first set condition, and determining a virtual predicted track which belongs to the obstacle and is over against the self through an interactive motion model, the obstacle motion parameter and the self motion parameter.
In the present invention, when the duration satisfies a first preset condition, determining a virtual predicted trajectory of the obstacle, which is directly facing the host vehicle, through an interactive motion model, the obstacle motion parameter, and the host vehicle motion parameter includes: when the duration meets a first preset condition, determining a virtual intersection point where the obstacle and a longitudinal axis where the self vehicle is located intersect through the speed direction of the obstacle; and constructing a virtual predicted track with the longitudinal coordinate same as the longitudinal coordinate of the self vehicle and the track end point as the virtual intersection point through the preset speed of the obstacle and the virtual intersection point.
Specifically, since the obstacle hardly has a longitudinal component velocity opposite to the direction of movement of the vehicle when the obstacle belongs to the crossing obstacle, the virtual predicted trajectory of the obstacle cannot be determined by the same embodiment as described above in which the virtual predicted trajectory of the obstacle facing the vehicle is determined by the interactive motion model, the obstacle motion parameter, and the vehicle motion parameter when the obstacle belongs to the facing obstacle.
At this time, the virtual intersection point of the self vehicle and the obstacle on the longitudinal axis of the self vehicle is directly determined as the longitudinal axis value of the obstacle at the current moment, namely, an auxiliary line is directly made in the direction of the speed of the obstacle at the current moment, and the intersection point of the auxiliary line and the longitudinal axis S axis of the self vehicle is the virtual intersection point. At this time, since the own vehicle will decelerate to stop at the virtual intersection, the time period taken until the own vehicle finally stops at the virtual intersection from the present time can be determined by the following equation (2):
wherein v is ego The speed of the current time of the self-vehicle belongs to a known quantity; s dis The distance from the vehicle to the virtual intersection point is a known quantity; is the deceleration of the own vehicle and belongs to an unknown quantity; t is the deceleration at which the vehicle travels and finally reaches the virtual intersectionThe length of time required to stop is unknown.
The values of the deceleration a and the time length t of the vehicle are obtained through calculation by the formula (2). Since the obstacle belongs to the crosscut obstacle, the obstacle hardly has a longitudinal component velocity opposite to the moving direction of the self-vehicle, and at the moment, in order to construct a virtual predicted track of the obstacle, the longitudinal coordinate of which is the same as the longitudinal coordinate of the self-vehicle, and the track end point of which is a virtual intersection point, a preset component velocity opposite to the moving direction of the self-vehicle of the crosscut obstacle needs to be preset. And constructing a virtual predicted track of the obstacle, wherein the longitudinal coordinate of the virtual predicted track is the same as the longitudinal coordinate of the vehicle, and the track end point is the virtual intersection point, based on the preset sub-speed and the virtual intersection point.
For example, as shown in fig. 7, a virtual intersection point of the host vehicle and the obstacle on a longitudinal axis of the host vehicle is directly determined as a longitudinal axis value of the obstacle at the current time, that is, an auxiliary line is directly drawn in a direction of a speed of the obstacle at the current time, and an intersection point of the auxiliary line and a longitudinal axis S axis is a virtual intersection point, as shown in P3 in fig. 7. The values of the deceleration a and the time length t of the vehicle are obtained through calculation by the two calculation formulas. Since the obstacle belongs to the cross obstacle, the obstacle hardly has a longitudinal component velocity opposite to the moving direction of the self vehicle, at this time, in order to construct a virtual predicted track of the obstacle, the longitudinal coordinate of which is the same as the longitudinal coordinate of the self vehicle, and the track end point is a virtual intersection point, a preset component velocity opposite to the moving direction of the self vehicle of the cross obstacle needs to be preset, such as V72_1 in fig. 7, and actually, the obstacle only has a V72 opposite to the axis L of the horizontal axis, and does not have the component velocity V72_1 opposite to the moving direction of the self vehicle, but has a virtual component velocity which is specified in advance by the system. And constructing a virtual predicted track of the obstacle, wherein the longitudinal coordinate of the obstacle is the same as the longitudinal coordinate of the vehicle, and the track end point of the obstacle is the virtual intersection point. The starting point of the virtual predicted track is determined by the value of the time length t and the preset minute speed obtained by calculating through the two calculation formulas, namely the longitudinal distance between the starting point P4 and the end point P3 in the virtual predicted track P4-P3 is the product of the preset minute speed and the time length t.
In this embodiment, the set duration may be set according to an actual application scenario, and is not specifically limited herein, for example, 10S, 5S, and the like are set, and the value of the preset minute speed may be reasonably set according to the actual application scenario, and is not specifically limited herein.
In the embodiment, since the s-t graph algorithm used by the downstream decision end in decision making is insensitive to the lateral variation, when the predicted trajectory intersecting the obstacle is obtained by the conventional technical means, the predicted trajectory almost has the lateral variation, and the downstream decision end cannot accurately avoid the obstacle of the predicted trajectory. Through the implementation mode of the invention, even though the actual track of the crosscut obstacle is almost transversely changed in reality, the virtual predicted track which is finally sent to the s-t graph algorithm of the downstream decision end is still longitudinally changed, and the s-t graph algorithm of the downstream decision end can perform deceleration obstacle avoidance decision more accurately and sensitively based on the received virtual predicted track which is longitudinally changed.
In the invention, when the type of the obstacle belongs to the equidirectional obstacle, the predicted track of the obstacle is determined through a uniform linear motion model and the obstacle motion parameters.
In this embodiment, when it is determined that the type of the obstacle to which the obstacle belongs is the equidirectional obstacle, the risk angle of the obstacle to the vehicle is low, and at this time, the predicted trajectory of the obstacle may be constructed directly by the speed, the speed direction, and the uniform linear motion model of the obstacle.
In the present invention, before determining whether there is an interaction intention between the obstacle and the host vehicle when the type of obstacle to which the obstacle belongs is an opposing obstacle, the method further includes: taking the position of the self-vehicle as an original point, and performing rasterization division of a preset size on the periphery of the self-vehicle to obtain a plurality of grid areas; determining the grid region types to which the grid regions belong according to the position relations between the grid regions and the own vehicle; when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle comprises the following steps: when the type of the obstacle belongs to the opposite obstacle, determining the grid area type of the obstacle; determining that the obstacle has an interaction intention with the host vehicle if the obstacle is within a grid area of a first grid area category; under the condition that the obstacle is in a grid area of a second grid area type, determining a first included angle between the speed direction of the obstacle and a transverse axis, and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis; determining that an interaction intention exists between the barrier and the vehicle under the condition that the first included angle is smaller than or equal to the second included angle, or determining that the interaction intention does not exist between the barrier and the vehicle under the condition that the first included angle is larger than the second included angle; determining that the obstacle has a target interaction intention with the host vehicle if the obstacle is within a grid area of a third grid area category; and under the condition that the obstacle and the self-vehicle have an interaction intention, determining the predicted track of the obstacle through a uniform-speed linear motion model and the obstacle motion parameters so as to sensitively respond to the motion direction of the obstacle.
In this embodiment, in order to further ensure the safety of the operation of the vehicle and improve the smoothness of the operation of the vehicle, the invention provides another embodiment of rasterizing the periphery of the vehicle body of the vehicle, specifically:
and rasterizing the periphery of the vehicle body in real time according to the coordinate position of the vehicle. Specifically, by setting the size of each grid in advance, the front of the vehicle and the periphery of the vehicle body are divided into 6 regions from the rear of the vehicle. One grid region surrounding the host vehicle is determined as a third grid region type, one grid region immediately in front of the grid region surrounding the host vehicle is determined as a first grid region type, and 4 grid regions on both left and right sides of the grid region surrounding the host vehicle and the grid region immediately in front of the grid region surrounding the host vehicle are determined as a second grid region type.
In the process of rasterizing the periphery of the vehicle body, the area behind the vehicle is not considered because in practical situations, obstacles behind the vehicle do not cause significant collision problems to the vehicle motion. Meanwhile, the reason why the grid is carried out around the self-vehicle is that the obstacles in the same grid area have common motion characteristics, the obstacles can be classified and considered uniformly, and the topological structures of the six areas can ensure the running stability and safety of the self-vehicle.
For example, as shown in fig. 8, the grid region 5 is a grid region surrounding the host vehicle, and thus the grid region 5 is determined as the third grid region category; in the figure, the grid region 2 is a grid region immediately in front of the grid region surrounding the own vehicle, and therefore the grid region 2 is determined as a first grid region type; in the figure, the grid regions 1, 3, 4, and 6 are left and right grid regions surrounding the host vehicle and right and left grid regions immediately in front of the grid region surrounding the host vehicle, and therefore the grid regions 1, 3, 4, and 6 are determined as the second grid region type.
Such a grid division of the periphery of the host vehicle is only for the type of opposing obstacles with a high degree of motion complexity. When the obstacle type to which the obstacle belongs is determined to be the opposite obstacle, the grid region type where the obstacle is located is further determined.
When the obstacle is in the grid area of the first grid area type, the obstacle is positioned right in front of the vehicle, the danger degree of the obstacle relative to the vehicle is high, and the interaction intention between the obstacle in the first grid area type and the vehicle is directly determined without further judgment. And then, determining a virtual predicted track which belongs to the barrier and is over against the self vehicle directly through an interactive motion model, the barrier motion parameter and the self vehicle motion parameter. It should be understood that, here, the specific implementation of determining the virtual predicted trajectory of the opposite obstacle to the host vehicle, which belongs to the obstacle, through the interactive motion model, the obstacle motion parameter and the host vehicle motion parameter is the same as the above-mentioned implementation of generating the virtual predicted trajectory of the opposite obstacle determined to have the interactive intention, and will not be described again here.
When the obstacle is in the grid area of the second grid area category, the obstacle is positioned on the left side and the right side of the vehicle, the danger degree of the obstacle relative to the vehicle is low, at this time, whether interaction intention exists between the obstacle and the vehicle is determined in the same manner as the above embodiment, that is, whether a first included angle between the speed direction of the obstacle and the horizontal axis exists is determined, a second included angle between the connecting line of the obstacle and the vehicle and the horizontal axis is determined, when the first included angle is smaller than or equal to the second included angle, the interaction intention exists between the obstacle and the vehicle is determined, and when the first included angle is larger than the second included angle, the interaction intention does not exist between the obstacle and the vehicle. And when the interactive intention exists, determining a virtual predicted track which belongs to the obstacle and is over against the self vehicle through an interactive motion model, the obstacle motion parameter and the self vehicle motion parameter. It should be understood that, here, the specific implementation of determining the virtual predicted trajectory of the opposite obstacle to the host vehicle, which belongs to the obstacle, through the interactive motion model, the obstacle motion parameter and the host vehicle motion parameter is the same as the above-mentioned implementation of generating the virtual predicted trajectory of the opposite obstacle determined to have the interactive intention, and will not be described again here. And when no interactive intention exists, determining a virtual predicted track of the obstacle through a non-interactive motion model and the obstacle motion parameters.
And under the condition that the barrier is in the grid area of the third grid area type, the barrier is in a position close to the periphery of the self-vehicle at the moment, the danger degree is very high, the target interaction intention, namely the interaction intention with higher danger degree, is determined to exist between the barrier and the self-vehicle at the moment, and the upstream directly passes through the constant-speed linear motion model and the speed of the barrier and faces to the predicted track for constructing the barrier. Since the obstacle is very close to the own vehicle at the moment, as long as the received predicted track has the moving direction towards the own vehicle, the downstream can immediately make a braking decision so as to avoid collision between the own vehicle and the obstacle, thereby achieving the purpose of sensitively responding to the moving direction of the obstacle.
In this embodiment, the size of each grid region may be set to be the same size, or may be set to be different sizes, which is not specifically limited herein, and meanwhile, the value of the size of each grid region may also be taken according to an actual application scenario, which is not specifically limited herein.
An embodiment of the present invention further provides a device 900 for trajectory prediction, as shown in fig. 9, the device 900 includes:
a speed direction determining unit 901, configured to determine a speed direction of the obstacle and a speed direction of the host vehicle according to the acquired obstacle motion parameter and the host vehicle motion parameter;
the obstacle type identification unit 902 is configured to determine the type of an obstacle to which the obstacle belongs according to a relationship between an included angle between a speed direction of the obstacle and a speed direction of the host vehicle and a preset angle range, where the preset angle range includes a first preset angle range, a second preset angle range, and a third preset angle range;
an interaction intention determining unit 903, configured to determine whether there is an interaction intention between the obstacle and the host vehicle when the obstacle type to which the obstacle belongs is an opposing obstacle;
a first virtual predicted trajectory generation unit 904, configured to determine, through an interactive motion model, the obstacle motion parameter, and the vehicle motion parameter, a virtual predicted trajectory that belongs to the obstacle and faces the vehicle when there is an interactive intention between the obstacle and the vehicle;
a second virtual predicted trajectory generating unit 905, configured to determine a virtual predicted trajectory of the obstacle through a non-interactive motion model and the obstacle motion parameter when there is no interaction intention between the obstacle and the host vehicle.
Optionally, the interaction intention determining unit 903 includes:
the included angle determining unit is used for determining a first included angle between the speed direction of the obstacle and a transverse axis and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis when the obstacle type of the obstacle is an opposite obstacle;
the interaction intention determining subunit is used for determining that an interaction intention exists between the obstacle and the own vehicle under the condition that the first included angle is smaller than or equal to the second included angle; and under the condition that the first included angle is larger than the second included angle, determining that no interaction intention exists between the barrier and the self-vehicle.
Optionally, the second virtual predicted trajectory generating unit 905 includes:
the component velocity determining unit is used for determining the component velocity of the obstacle opposite to the moving direction of the self vehicle according to the obstacle moving parameter;
and the second virtual predicted track generating subunit is used for constructing a virtual predicted track of the obstacle, wherein the vertical coordinate of the virtual predicted track is the same as the vertical coordinate of the position of the obstacle at the current moment, and the starting point of the virtual predicted track is located at the position of the obstacle at the current moment, according to the component speed.
Optionally, the apparatus 900 further comprises: a third virtual predicted trajectory generation unit; the third virtual predicted trajectory generation unit includes:
the duration determining unit is used for determining the duration required by the obstacle to cut into the front of the vehicle according to the obstacle motion parameters when the type of the obstacle to which the obstacle belongs is a transverse-cutting obstacle;
a third virtual predicted track generation subunit, configured to determine, when the duration satisfies a first preset condition, a virtual predicted track that belongs to the obstacle and faces the host vehicle, through an interactive motion model, the obstacle motion parameter, and the host vehicle motion parameter; and when the duration does not meet a first preset condition, determining the predicted track of the obstacle through a uniform linear motion model and the obstacle motion parameters.
Optionally, the apparatus 900 further comprises:
the grid dividing unit is used for carrying out grid division on the periphery of the self-vehicle by a preset size by taking the position of the self-vehicle as an origin to obtain a plurality of grid areas;
the grid region classification unit is used for determining grid region categories to which the grid regions belong according to the position relations between the grid regions and the self-vehicle;
the interaction intention determining unit 903 includes: the first interaction intention determining unit is used for determining the grid area type of the obstacle when the obstacle type of the obstacle is an opposite obstacle; determining that the obstacle has an interaction intention with the host vehicle if the obstacle is within a grid area of a first grid area category; under the condition that the obstacle is in a grid area of a second grid area type, determining a first included angle between the speed direction of the obstacle and a transverse axis, and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis; determining that an interaction intention exists between the barrier and the vehicle under the condition that the first included angle is smaller than or equal to the second included angle, or determining that the interaction intention does not exist between the barrier and the vehicle under the condition that the first included angle is larger than the second included angle; determining that the obstacle has a target interaction intention with the host vehicle if the obstacle is within a grid area of a third grid area category;
and the fourth virtual predicted track generating unit is used for determining the predicted track of the obstacle through a uniform-speed linear motion model and the obstacle motion parameters under the condition that the obstacle and the self vehicle have an interaction intention so as to sensitively respond to the motion direction of the obstacle.
Optionally, the first virtual predicted trajectory generating unit 904 includes:
a first partial speed determination unit, configured to determine a partial speed of the obstacle opposite to a moving direction of the host vehicle through the obstacle motion parameter when there is an interaction intention between the obstacle and the host vehicle;
a first virtual intersection point determining unit, configured to determine a virtual intersection point of the vehicle and the obstacle on a longitudinal axis where the vehicle is located, according to the component speed, a speed of the vehicle at a current time, and a distance between the obstacle and the vehicle in the longitudinal direction;
and the first virtual predicted track generation subunit is used for constructing a virtual predicted track with the longitudinal coordinate same as the longitudinal coordinate of the vehicle and the track end point as the virtual intersection point through the branch speed and the virtual intersection point.
Optionally, the third virtual predicted trajectory generation subunit includes:
the second virtual intersection point determining unit is used for determining a virtual intersection point where the barrier and the longitudinal axis where the self vehicle is located intersect through the speed direction of the barrier when the duration meets a first preset condition;
and the fifth virtual predicted track generation unit is used for constructing a virtual predicted track with the longitudinal coordinate same as the longitudinal coordinate of the self-vehicle and the track end point as the virtual intersection point through the preset speed of the obstacle and the virtual intersection point.
The embodiment of the invention also provides electronic equipment which comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus; a memory for storing a computer program; and the processor is used for realizing the steps of the track prediction method when executing the program stored in the memory.
Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps in the method for trajectory prediction.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (9)
1. A method of trajectory prediction, the method comprising:
determining the speed direction of the obstacle and the speed direction of the self-vehicle according to the acquired obstacle motion parameters and the self-vehicle motion parameters;
determining the type of the obstacle to which the obstacle belongs according to the relation between the included angle between the speed direction of the obstacle and the speed direction of the vehicle and a preset angle range, wherein the preset angle range comprises a first preset angle range, a second preset angle range and a third preset angle range;
when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle;
under the condition that the barrier and the self vehicle have interactive intention, determining a virtual predicted track which belongs to the barrier and is over against the self vehicle through an interactive motion model, the barrier motion parameter and the self vehicle motion parameter;
under the condition that no interaction intention exists between the obstacle and the self-vehicle, determining a virtual predicted track of the obstacle through a non-interactive motion model and the obstacle motion parameters;
wherein, in the case that there is an interaction intention between the obstacle and the host vehicle, determining a virtual predicted trajectory belonging to the obstacle and facing the host vehicle through an interaction motion model, the obstacle motion parameter, and the host vehicle motion parameter includes: determining the partial speed of the obstacle opposite to the moving direction of the self vehicle through the obstacle moving parameter under the condition that the obstacle and the self vehicle have the interaction intention; determining a virtual intersection point of the self vehicle and the obstacle on a longitudinal axis of the self vehicle according to the partial speed, the speed of the self vehicle at the current moment and the distance between the obstacle and the self vehicle in the longitudinal direction; and constructing a virtual prediction track with the longitudinal coordinate same as the longitudinal coordinate of the self vehicle and the track end point as the virtual intersection point through the partial speed and the virtual intersection point.
2. The method for trajectory prediction according to claim 1, wherein the determining whether there is an interaction intention between the obstacle and the host vehicle when the obstacle type to which the obstacle belongs is an opposing obstacle comprises:
when the type of the barrier to which the barrier belongs is an opposite barrier, determining a first included angle between the speed direction of the barrier and a transverse shaft, and determining a second included angle between a connecting line of the barrier and the vehicle and the transverse shaft;
determining that an interaction intention exists between the barrier and the self-vehicle under the condition that the first included angle is smaller than or equal to the second included angle;
and under the condition that the first included angle is larger than the second included angle, determining that no interaction intention exists between the barrier and the self-vehicle.
3. The method of claim 1, wherein the determining the virtual predicted trajectory of the obstacle through the non-interactive motion model and the obstacle motion parameter without the intention of interaction between the obstacle and the host vehicle comprises:
determining the component speed of the obstacle opposite to the movement direction of the self vehicle through the obstacle movement parameters;
and constructing a virtual predicted track of the obstacle, wherein the longitudinal coordinate of the virtual predicted track is the same as that of the position of the obstacle at the current moment, and the starting point of the virtual predicted track is located at the position of the obstacle at the current moment.
4. The method for predicting the track according to claim 1, wherein when the type of the obstacle to which the obstacle belongs is a transverse obstacle, the time length required for the obstacle to cut into the front of the vehicle is determined according to the obstacle motion parameters;
when the duration meets a first preset condition, determining a virtual predicted track which belongs to the barrier and is over against the self-vehicle through an interactive motion model, the barrier motion parameter and the self-vehicle motion parameter;
and when the duration does not meet a first preset condition, determining the predicted track of the obstacle through a uniform linear motion model and the obstacle motion parameter.
5. The method of claim 3, wherein before determining whether there is an interaction intention between the obstacle and the host vehicle when the type of obstacle to which the obstacle belongs is an oncoming obstacle, the method further comprises:
taking the position of the self-vehicle as an original point, and performing rasterization division of a preset size on the periphery of the self-vehicle to obtain a plurality of grid areas;
determining grid region types to which the grid regions belong according to the position relations between the grid regions and the vehicle;
when the type of the obstacle to which the obstacle belongs is an opposite obstacle, determining whether an interaction intention exists between the obstacle and the vehicle comprises the following steps:
when the type of the obstacle belongs to the opposite obstacle, determining the grid area type of the obstacle; determining that the obstacle has an interaction intention with the host vehicle if the obstacle is within a grid area of a first grid area category; under the condition that the obstacle is in a grid area of a second grid area type, determining a first included angle between the speed direction of the obstacle and a transverse axis, and determining a second included angle between a connecting line of the obstacle and the self vehicle and the transverse axis; determining that an interaction intention exists between the obstacle and the self-vehicle under the condition that the first included angle is smaller than or equal to the second included angle, or determining that no interaction intention exists between the obstacle and the self-vehicle under the condition that the first included angle is larger than the second included angle; determining that the obstacle has a target interaction intention with the host vehicle if the obstacle is within a grid area of a third grid area category;
and under the condition that the obstacle and the self-vehicle have an interaction intention, determining the predicted track of the obstacle through a uniform-speed linear motion model and the obstacle motion parameters so as to sensitively respond to the motion direction of the obstacle.
6. The method for predicting the track according to claim 4, wherein when the duration satisfies a first preset condition, determining a virtual predicted track, belonging to the obstacle, facing the host vehicle through an interactive motion model, the obstacle motion parameter and the host vehicle motion parameter comprises:
when the duration meets a first preset condition, determining a virtual intersection point where the obstacle and a longitudinal axis where the self vehicle is located intersect through the speed direction of the obstacle;
and constructing a virtual predicted track with the longitudinal coordinate same as the longitudinal coordinate of the self vehicle and the track end point as the virtual intersection point through the preset speed of the obstacle and the virtual intersection point.
7. An apparatus for trajectory prediction, the apparatus comprising:
the speed direction determining unit is used for determining the speed direction of the obstacle and the speed direction of the self-vehicle according to the acquired obstacle motion parameters and the self-vehicle motion parameters;
the obstacle type identification unit is used for determining the type of the obstacle to which the obstacle belongs according to the relation between the included angle between the speed direction of the obstacle and the speed direction of the vehicle and a preset angle range, wherein the preset angle range comprises a first preset angle range, a second preset angle range and a third preset angle range;
an interaction intention determining unit, configured to determine whether there is an interaction intention between the obstacle and the host vehicle when the obstacle type to which the obstacle belongs is an opposing obstacle;
a first virtual predicted track generation unit, configured to determine, through an interactive motion model, the obstacle motion parameter, and the vehicle motion parameter, a virtual predicted track that belongs to the obstacle and faces the vehicle when there is an interactive intention between the obstacle and the vehicle;
a second virtual predicted track generation unit, configured to determine a virtual predicted track of the obstacle through a non-interactive motion model and the obstacle motion parameter when there is no interactive intention between the obstacle and the host vehicle;
wherein the first virtual predicted trajectory generation unit includes: a first sub-speed determining unit, configured to determine, through the obstacle motion parameter, a sub-speed of the obstacle opposite to a motion direction of the host vehicle when there is an interaction intention between the obstacle and the host vehicle; a first virtual intersection point determining unit, configured to determine a virtual intersection point of the vehicle and the obstacle on a longitudinal axis where the vehicle is located, according to the component speed, a speed of the vehicle at a current time, and a distance between the obstacle and the vehicle in the longitudinal direction; and the first virtual predicted track generation subunit is used for constructing a virtual predicted track with the longitudinal coordinate same as the longitudinal coordinate of the vehicle and the track end point as the virtual intersection point through the branch speed and the virtual intersection point.
8. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for implementing the steps of a method of trajectory prediction according to any one of claims 1 to 6 when executing a program stored in a memory.
9. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program when executed by a processor implementing the steps in a method of trajectory prediction according to any of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211140393.7A CN115214724B (en) | 2022-09-20 | 2022-09-20 | Trajectory prediction method and apparatus, electronic device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211140393.7A CN115214724B (en) | 2022-09-20 | 2022-09-20 | Trajectory prediction method and apparatus, electronic device and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115214724A CN115214724A (en) | 2022-10-21 |
CN115214724B true CN115214724B (en) | 2022-12-09 |
Family
ID=83616968
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211140393.7A Active CN115214724B (en) | 2022-09-20 | 2022-09-20 | Trajectory prediction method and apparatus, electronic device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115214724B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115900697B (en) * | 2022-10-28 | 2024-03-29 | 北京百度网讯科技有限公司 | Object motion trail information processing method, electronic equipment and automatic driving vehicle |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007265451A (en) * | 2007-07-17 | 2007-10-11 | Matsushita Electric Works Ltd | Obstacle detector for vehicle |
CN111079721A (en) * | 2020-03-23 | 2020-04-28 | 北京三快在线科技有限公司 | Method and device for predicting track of obstacle |
WO2021135728A1 (en) * | 2019-12-30 | 2021-07-08 | 郑州宇通客车股份有限公司 | Determination method and device for collision prediction of autonomous vehicle |
CN113264066A (en) * | 2021-06-03 | 2021-08-17 | 阿波罗智能技术(北京)有限公司 | Obstacle trajectory prediction method and device, automatic driving vehicle and road side equipment |
CN114312698A (en) * | 2022-01-24 | 2022-04-12 | 岚图汽车科技有限公司 | Vehicle braking method, device, equipment and readable storage medium |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110550029B (en) * | 2019-08-12 | 2021-02-09 | 华为技术有限公司 | Obstacle avoiding method and device |
-
2022
- 2022-09-20 CN CN202211140393.7A patent/CN115214724B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007265451A (en) * | 2007-07-17 | 2007-10-11 | Matsushita Electric Works Ltd | Obstacle detector for vehicle |
WO2021135728A1 (en) * | 2019-12-30 | 2021-07-08 | 郑州宇通客车股份有限公司 | Determination method and device for collision prediction of autonomous vehicle |
CN111079721A (en) * | 2020-03-23 | 2020-04-28 | 北京三快在线科技有限公司 | Method and device for predicting track of obstacle |
CN113264066A (en) * | 2021-06-03 | 2021-08-17 | 阿波罗智能技术(北京)有限公司 | Obstacle trajectory prediction method and device, automatic driving vehicle and road side equipment |
CN114312698A (en) * | 2022-01-24 | 2022-04-12 | 岚图汽车科技有限公司 | Vehicle braking method, device, equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN115214724A (en) | 2022-10-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110834631A (en) | Pedestrian avoiding method and device, vehicle and storage medium | |
JP6468261B2 (en) | Automated driving system | |
JP7145815B2 (en) | electronic controller | |
WO2021196041A1 (en) | Selection method for key target, apparatus, and system | |
CN115214724B (en) | Trajectory prediction method and apparatus, electronic device and storage medium | |
WO2021054051A1 (en) | Electronic control device | |
US20220203971A1 (en) | Vehicle control method | |
Park et al. | A vehicle speed harmonization strategy for minimizing inter-vehicle crash risks | |
CN106218612A (en) | A kind of method of vehicle safety travel, device and terminal | |
CN102298847A (en) | Publishing system and method for linkage of road network information | |
CN114734996A (en) | Automatic driving decision method, device, vehicle and storage medium | |
Weisswange et al. | Intelligent traffic flow assist: Optimized highway driving using conditional behavior prediction | |
CN115223131A (en) | Adaptive cruise following target vehicle detection method and device and automobile | |
CN113538909A (en) | Traffic incident prediction method and device for automatic driving vehicle | |
CN113432615A (en) | Detection method and system based on multi-sensor fusion drivable area and vehicle | |
CN112365730A (en) | Automatic driving method, device, equipment, storage medium and vehicle | |
CN113682300B (en) | Decision method, device, equipment and medium for avoiding obstacle | |
Ding et al. | Research on safety lane change warning method based on potential angle collision point | |
CN114371015B (en) | Automatic driving test method, automatic driving test device, computer equipment and storage medium | |
CN115214722A (en) | Automatic driving method, device, electronic equipment, storage medium and vehicle | |
CN115270999B (en) | Obstacle risk grade classification method and device, storage medium and vehicle | |
CN111176285A (en) | Method and device for planning travel path, vehicle and readable storage medium | |
CN116572994B (en) | Vehicle speed planning method, device and computer readable medium | |
CN117184062B (en) | Vehicle control method and device, vehicle and storage medium | |
Lin et al. | Adaptive trajectory generation of autonomous vehicle in urban environments |
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 |