WO2022104790A1

WO2022104790A1 - Travel trajectory determining method and apparatus, computer device, and storage medium

Info

Publication number: WO2022104790A1
Application number: PCT/CN2020/130873
Authority: WO
Inventors: 郑少华; 林川
Original assignee: 深圳元戎启行科技有限公司
Priority date: 2020-11-23
Filing date: 2020-11-23
Publication date: 2022-05-27
Also published as: CN114945959B; CN114945959A

Abstract

A travel trajectory determining method, comprising: obtaining vehicle travel information of vehicles to be corrected in a set of vehicles to be corrected; determining abnormal state indexes of said vehicles on the basis of the vehicle travel information of said vehicles; obtaining environmental information of said vehicles in said set of vehicles; and according to the order of the abnormal state indexes in an abnormal state index sequence, on the basis of the environmental information of said vehicles, respectively determining predicted travel trajectories of said vehicles after being corrected, the abnormal state index sequence being formed by ordering from weak to strong according to abnormal degrees represented by the abnormal state indexes.

Description

Driving trajectory determination method, device, computer equipment and storage medium

technical field

The present application relates to the technical field of vehicle driving, and in particular, to a driving trajectory determination method, device, computer device and computer-readable storage medium.

Background technique

Every year, people are injured or disabled due to traffic accidents, and cause great economic losses. Autonomous driving technology can improve driving safety, improve the efficiency of the entire transportation system, and save time for users.

The key modules of autonomous driving technology mainly include positioning, perception, prediction, planning, decision-making and control. The decision-making planning relies on the environmental information of positioning, perception and prediction, and comprehensively analyzes the above information to give comfortable and safe autonomous driving planning actions. The level of autonomous driving planning technology represents the ability of autonomous vehicles to deal with complex traffic scenarios. However, the current predicted driving trajectories generated for vehicles are inaccurate.

SUMMARY OF THE INVENTION

Various embodiments of the present application provide a driving trajectory determination method, apparatus, computer device, and storage medium.

A method for determining a driving trajectory, the method comprising:

Obtain vehicle driving information of each vehicle to be corrected in the vehicle set to be corrected;

Determine the abnormal state index of each to-be-corrected vehicle based on the vehicle driving information of each to-be-corrected vehicle;

obtaining the environmental information of each vehicle to be corrected in the set of vehicles to be corrected; and

According to the order of the abnormal state indices in the abnormal state index sequence, the corrected predicted driving trajectory of each to-be-corrected vehicle is determined based on the environmental information of each of the to-be-corrected vehicles; the abnormal state index sequence is represented by each abnormal state index The degree of anomaly is arranged from weak to strong.

A driving trajectory determination device, the device includes:

a first obtaining module, configured to obtain vehicle driving information of each vehicle to be corrected in the set of vehicles to be corrected;

an abnormal state index determination module, configured to determine the abnormal state index of each to-be-corrected vehicle based on the vehicle driving information of each to-be-corrected vehicle;

a second obtaining module, configured to obtain the environmental information of each vehicle to be corrected in the set of vehicles to be corrected;

a correction module, configured to determine the corrected predicted driving trajectory of each vehicle to be corrected based on the environmental information of each vehicle to be corrected according to the order of the abnormal state index in the abnormal state index sequence; The degree of anomaly represented by the state index is arranged from weak to strong.

A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements each step in the method embodiment when the processor executes the computer program.

A computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements each step in this embodiment of the method.

In the above-mentioned driving trajectory determination method, device, computer equipment and storage medium, in normal driving traffic, vehicles with a high degree of abnormality are generally the source of changes in traffic flow. For example, a stationary vehicle in front will cause the vehicle behind it to slow down and change lanes. Give way, the abnormal state will be gradually transmitted from the source over time to affect the entire traffic flow, that is, vehicles with low abnormality will be affected by vehicles with high abnormality, and vehicles with high abnormality will be affected by vehicles with low abnormality. The influence is small. Based on this principle, this embodiment determines the abnormal state index of each vehicle to be corrected based on the vehicle driving information of the vehicle to be corrected, and based on the environmental information of each vehicle to be corrected, the predicted trajectory is corrected from the vehicle with a low degree of abnormality to advance The impact of abnormal conditions in the traffic flow on the normal driving vehicles is predicted, so as to determine the corrected predicted driving trajectory of the vehicles in the vehicle set to be corrected, and improve the accuracy of the predicted driving trajectory of the vehicle.

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

FIG. 1 is an application environment diagram of a method for determining a driving trajectory in one embodiment.

FIG. 2 is a schematic flowchart of a method for determining a driving trajectory in an embodiment.

FIG. 3 is a schematic flowchart of determining the corrected predicted travel trajectory of each vehicle to be corrected based on the environmental information of each vehicle to be corrected in one embodiment.

FIG. 4 is a schematic diagram of a preset distance range in one embodiment.

FIG. 5 is a schematic flowchart of a method for determining a driving trajectory in another embodiment.

FIG. 6 is a schematic flowchart of determining a revised predicted driving trajectory based on environmental information in one embodiment.

FIG. 7 is a structural block diagram of an apparatus for determining a driving trajectory in an embodiment.

FIG. 8 is a diagram of the internal structure of a computer device in one embodiment.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

The driving trajectory determination method provided by the present application can be applied to the application environment shown in FIG. 1 . At least two vehicles 110 to be corrected are included in FIG. 1 . The methods in the embodiments of the present application can be applied to any vehicle to be corrected, and the vehicle to be corrected has a computer device, and the computer device may specifically be a vehicle-mounted device. The in-vehicle device can be a personal computer, a laptop, a smartphone, a tablet, and the like. The vehicle to be corrected 110 may further include a camera, a radar, etc., and both the camera and the radar of the vehicle to be corrected may be connected to the in-vehicle device. The set of vehicles to be corrected in the scene can be known through cameras or radars. The vehicle A to be corrected is taken as an example for description, and the vehicle A to be corrected has a computer device installed therein. The computer device acquires vehicle travel information of each vehicle to be corrected in the set of vehicles to be corrected. That is, the vehicle travel information of the vehicle A to be corrected, the vehicle travel information of the vehicle B to be corrected, and the vehicle travel information of the vehicle C to be corrected. The computer device determines the abnormal state index of each vehicle to be corrected based on the vehicle travel information of each vehicle to be corrected. That is, the abnormal state index of the vehicle A to be corrected is determined based on the vehicle travel information of the vehicle A to be corrected. The abnormal state index of the vehicle B to be corrected is determined based on the vehicle travel information of the vehicle B to be corrected. The abnormal state index of the vehicle C to be corrected is determined based on the vehicle travel information of the vehicle C to be corrected. Obtain the environmental information of each vehicle to be corrected in the set of vehicles to be corrected, that is, the environmental information of vehicle A to be corrected, the environmental information of vehicle B to be corrected, and the environmental information of vehicle C to be corrected. According to the order of the abnormal state indices in the abnormal state index sequence, the corrected predicted travel trajectory of each to-be-corrected vehicle is determined based on the environmental information of each to-be-corrected vehicle. That is, when the abnormal state index is arranged as the vehicle C to be corrected, the vehicle B to be corrected and the vehicle A to be corrected, then first determine the corrected predicted running trajectory of the to-be-corrected vehicle C, and then determine the corrected predicted running trajectory of the to-be-corrected vehicle B. . Next, determine the corrected predicted travel trajectory of the vehicle A to be corrected.

In one embodiment, currently in the field of automatic driving, the trajectory method of predicting the surrounding vehicles has more tendency and the accuracy of the individual, and the evaluation method is also the accuracy of the tendency and the individual, which cannot be compared with the surrounding vehicles. interact. The traditional method is based on the interaction between multiple vehicles, and gives the corresponding evaluation function and game matrix, but fails to effectively consider the basic map information, traffic rules and dynamic models between different vehicles, so the accuracy of the generated trajectory is difficult to measure. control. Therefore, in order to solve the problem of the traditional method, as shown in FIG. 2 , which is a schematic flowchart of a method for determining a driving trajectory in one embodiment, a method for determining a driving trajectory is provided. The following steps:

Step 202: Obtain vehicle driving information of each vehicle to be corrected in the vehicle set to be corrected.

Wherein, the vehicle to be corrected refers to the vehicle to be corrected for the predicted driving trajectory. The vehicle to be corrected may be an autonomous vehicle or a non-autonomous vehicle. When the vehicle to be corrected is an automatic driving vehicle, the predicted driving trajectory is used to instruct the vehicle to be corrected to automatically drive according to the predicted driving trajectory. When the vehicle to be corrected is a non-autonomous driving vehicle, the predicted driving trajectory is used to guide the driver to execute the vehicle driving according to the predicted driving trajectory. The set of vehicles to be corrected refers to a set of vehicles in a scene that have not been corrected for predicted driving trajectories. Specifically, the scene may be a scene in which the vehicle is driving on the road, a scene in which the vehicle is parked in a parking lot, or a scene in a traffic jam, which is not limited thereto.

Specifically, the computer device performs driving analysis on each vehicle to be corrected in the set of vehicles to be corrected, and obtains vehicle driving information of each vehicle to be corrected. The vehicle driving information refers to the information generated by the vehicle during the driving process. For example, the vehicle traveling information refers to the traveling speed of the vehicle, the lateral distance of the vehicle from the center line of the lane, the lateral speed of the vehicle, the orientation of the vehicle, and the like is not limited thereto.

Step 204: Determine the abnormal state index of each vehicle to be corrected based on the vehicle travel information of each vehicle to be corrected.

Among them, the abnormal state index is used to represent the abnormal degree of the vehicle. For example, the abnormal state may include that the vehicle suddenly stops on the road, the vehicle is driving on the line, the change rate of the vehicle's driving trajectory is large, and the vehicle suddenly loses control, etc., but is not limited thereto.

Specifically, the computer device determines the abnormal state index of each vehicle with correction based on the vehicle travel information of each vehicle to be corrected. Optionally, the computer device converts the driving information of each vehicle to be corrected into dimensionless units to obtain the abnormal state index of each vehicle to be corrected. Optionally, the computer device determines the abnormal state index of each vehicle to be corrected according to the comparison result of the vehicle driving information of each vehicle to be corrected and the vehicle driving reference value. Optionally, the computer device determines the abnormal state index of each vehicle to be corrected based on at least one of the lateral distance from the centerline of the lane, the current lateral speed of the vehicle, the current driving speed, and the current vehicle orientation.

Step 206: Obtain the environmental information of each vehicle to be corrected in the vehicle set to be corrected.

Specifically, the computer device acquires the environmental information of each vehicle to be corrected in the set of vehicles to be corrected. The environmental information refers to the information of the surrounding environment of the vehicle to be corrected. The environmental information specifically includes environmental vehicle information and environmental area information. The environmental vehicle information refers to information of vehicles around the vehicle to be corrected. The environmental vehicle information may include travel information of the environmental vehicle and the like. For example, the environmental vehicle information may include the predicted driving trajectory of the environmental vehicle, the number of environmental vehicles, and the like. The environmental area information refers to the surrounding area information of the vehicle to be corrected. For example, there are area 1, area 2, area 3, etc. around the vehicle to be corrected.

Step 208 , according to the order of the abnormal state indices in the abnormal state index sequence, determine the corrected predicted driving trajectory of each to-be-corrected vehicle based on the environmental information of each to-be-corrected vehicle; The degree is arranged from weak to strong.

Among them, the abnormal state index sequence is formed according to the abnormal degree represented by each abnormal state from weak to strong, that is, the index with weak abnormal degree is ranked in the front part of the abnormal state index sequence, and the index with strong abnormal degree is ranked in the abnormal state index sequence. in the back part. The degree of abnormality refers to the degree of abnormality of the vehicle with respect to the normal state of traveling along the lane. For example, the abnormality of the sudden loss of control of the vehicle is greater than the abnormality of the sudden shutdown of the vehicle. The degree of abnormality can be visually represented by the abnormal state index. For example, according to the setting, the larger the abnormal state index is, the higher the abnormality degree is; or it can be set that the abnormal state index is small and the abnormality degree is high. The relationship between the abnormality degree and the abnormality state index depends on the setting of the relational expression. The predicted driving trajectory refers to the vehicle driving trajectory in the future period of time of the vehicle to be corrected. The predicted travel trajectory includes the trajectory points of the traveled path.

Specifically, the revised predicted travel trajectory is the updated predicted travel trajectory. The corrected predicted travel trajectory may be corrected on the original predicted travel trajectory, or may be re-predicted based on the environmental information of each corrected vehicle. According to the order of the abnormal state indices in the abnormal state index sequence, the computer device determines the corrected predicted travel trajectory of each to-be-corrected vehicle based on the environmental information of each to-be-corrected vehicle. For example, the vehicle A to be corrected corresponds to the abnormal state index 1, the vehicle to be corrected B corresponds to the abnormal state index 2, and the vehicle C to be corrected corresponds to the abnormal state index 3, and the abnormality degree represented is arranged as abnormal state index 3<abnormal state index 2< If the abnormal state index is 1, then the order of the vehicles to be corrected is vehicle C to be corrected → vehicle B to be corrected → vehicle A to be corrected.

Optionally, the environmental information includes environmental area information and environmental vehicle information. The computer device determines a set of drivable areas of the vehicle to be revised based on the environmental area information of each vehicle to be revised; determines a target driving area of the vehicle to be revised from the set of drivable areas based on the environmental vehicle information; determines the correction of the vehicle to be revised based on the target driving area the predicted driving trajectory.

In the method for determining the driving trajectory in this embodiment, in the normal driving traffic flow, vehicles with a high degree of abnormality are generally the source of changes in the traffic flow. For example, a stationary vehicle in front will cause the vehicle behind it to slow down and change lanes. The abnormal state It will be gradually transmitted from the source over time to affect the entire traffic flow, that is, vehicles with low abnormality will be affected by vehicles with high abnormality, while vehicles with high abnormality will be less affected by vehicles with low abnormality. Based on this principle, the embodiment determines the abnormal state index of each vehicle to be corrected based on the vehicle driving information of the vehicle to be corrected, and corrects the predicted trajectory from the vehicle with a low degree of abnormality based on the environmental information of each vehicle to be corrected, so as to predict the abnormality in the traffic flow in advance. The influence of the situation on the normal driving vehicle is determined, so as to determine the corrected predicted driving trajectory of the vehicle in the vehicle set to be corrected, and improve the accuracy of the predicted driving trajectory of the vehicle.

In one embodiment, the vehicle driving information includes the lateral distance from the centerline of the lane, the current lateral speed of the vehicle, the current driving speed and the current vehicle orientation. Determining the abnormal state index of each vehicle to be corrected based on the vehicle driving information of each vehicle to be corrected includes: determining the lateral distance deviation index of each vehicle to be corrected based on the lateral distance of each vehicle to be corrected from the centerline of the lane where it is located; The current vehicle lateral speed of the vehicle determines the lateral speed deviation index of each vehicle to be corrected; the driving speed deviation index of each vehicle to be corrected is determined based on the current driving speed of each vehicle to be corrected; based on the current vehicle orientation of each vehicle to be corrected and the vehicle to be corrected The orientation deviation index of each vehicle to be corrected is determined by the centerline orientation of the lane where it is located; the abnormal state index of each vehicle to be corrected is determined based on the lateral distance deviation index, lateral speed deviation index, driving speed deviation index and orientation deviation index of each vehicle to be corrected.

The lateral distance from the center line of the lane where the vehicle is located refers to the lateral distance of the vehicle from the center line of the lane where it is located. Specifically, the vehicle center, the center of gravity of the vehicle, or the center of mass of the vehicle may be the lateral distance from the center line of the lane where the vehicle is located. Longitudinal refers to the lane direction, and lateral refers to the direction perpendicular to the lane direction. The current lateral speed of the vehicle refers to the speed of the vehicle in the direction perpendicular to the lane. The current travel speed may specifically refer to the current travel speed, that is, a scalar. For example, the current traveling speed is 60 km/h (kilometer per hour, 60 kilometers per hour), etc., but not limited to this. The current vehicle heading refers to the deviation angle of the vehicle relative to the centerline of its own lane. For example, if the current vehicle orientation is 60 degrees, it means that the vehicle to be corrected may be ready to turn around and so on.

Specifically, the computer device determines the lateral distance deviation index of each vehicle to be corrected based on the lateral distance of the vehicle to be corrected from the lane centerline. The computer device determines a lateral speed deviation index for each vehicle to be corrected based on the current vehicle lateral speed of each vehicle to be corrected. Optionally, in the absence of abnormality, the current lateral speed of the vehicle should be 0. The lateral speed deviation index of each vehicle to be corrected can then be determined directly based on the current vehicle lateral speed of each vehicle to be corrected. The computer device determines the travel speed deviation index of each vehicle to be corrected based on the current travel speed of each vehicle to be corrected. Optionally, the faster the current running speed of the vehicle to be corrected is, the higher the degree of abnormality of the vehicle to be corrected is indicated. The computer device determines the abnormal state index of each vehicle to be corrected based on the sum of the lateral distance deviation index, lateral speed deviation index, travel speed deviation index and heading deviation index of each to-be-corrected vehicle. Alternatively, the computer equipment multiplies the corresponding weights based on the lateral distance deviation index, lateral speed deviation index, driving speed deviation index and orientation deviation index of each vehicle to be corrected, and then sums them up to obtain the abnormal state index of each vehicle to be corrected.

For example, the lateral distance deviation index:

V _{lat_dis} =w _{lat_dis} *|lateral _cur |

where lateral _cur is the lateral distance of the vehicle to be corrected from the center line of the lane where it is located, and w _latdis is the weight.

Lateral Velocity Deviation Index:

V _{lat_speed} = w _{lat_speed} *|lat_speed|

where lat _speed is the current lateral speed of the vehicle, and w _{lat_speed} is the weight.

Driving Speed Deviation Index:

Where lon_speed _cur is the current driving speed, speed _limit is the current lane speed limit, w _below and w _over are the weights.

Towards Bias Index:

V _heading =w _heading *|heading _cur -heading _lane |

Where heading _cur is the current vehicle heading, heading _lane is the heading of the lane center point closest to the center of gravity of the vehicle, and w _heading is the weight.

Then the abnormal state index of the vehicle can be expressed as:

V=V _{lat_dis} +V _{lat_speed} +V _{lon_speed} +V _heading

In the method for determining the driving trajectory in this embodiment, in the normal driving state, the vehicle generally drives along the lane line and complies with the traffic regulations; since the lateral distance from the center line of the vehicle, the current lateral speed of the vehicle and the current vehicle orientation are all It can be used to characterize whether the vehicle has a tendency to change lanes, whether it will lose control, etc. Therefore, the lateral distance from the center line of the vehicle can be considered as part of the abnormal state, and the lateral distance deviation index, lateral speed deviation index and orientation deviation index can be obtained; current; The driving speed can be used to characterize whether the vehicle is driving faster or slower, etc. Therefore, the current driving speed is also used to characterize whether the vehicle is driving too fast or too slow, etc., so as to accurately determine the abnormal degree of the vehicle, so that the vehicle's abnormality can be determined in sequence. The revised predicted driving trajectory.

In one embodiment, the determining the driving speed deviation index of each vehicle to be corrected based on the current driving speed of each vehicle to be corrected includes: acquiring the lane speed limit corresponding to the lane where each vehicle to be corrected is located; based on the current driving speed and The lane speed limit determines the speed deviation value of each vehicle to be corrected; and determines the running speed deviation index of each vehicle to be corrected based on the speed deviation value.

The lane speed limit refers to the speed limit value of the lane. The lane limit may be the speed limit value of the road section, or the speed limit value of the lane itself. The speed deviation value is used to indicate the deviation between the current driving speed and the lane speed limit. The speed deviation value may be the speed difference value between the current driving speed and the lane speed limit, or may be the difference ratio value between the current driving speed and the lane speed limit, etc., but not limited to this.

Specifically, the computer device acquires the lane speed limit corresponding to the lane where each vehicle to be corrected is located from the environmental semantic map. The computer device determines the speed deviation value of each vehicle to be corrected based on the current travel speed and the lane speed limit. Alternatively, the computing device may determine the speed offset value based on the difference between the current travel speed and the lane speed limit. Alternatively, the computing device may determine the speed offset value based on the absolute value of the difference between the current travel speed and the lane speed limit. Alternatively, based on the difference between the current driving speed and the lane speed limit, the computer may use the ratio of the difference to the lane speed limit as the speed deviation value of the vehicle to be corrected.

For example, the driving speed deviation index:

In the method for determining the driving trajectory in this embodiment, the vehicle is driving on the lane. Generally, there are different speed limits on different road sections. Therefore, the speed deviation value of each vehicle to be corrected can be determined based on the current driving speed of the vehicle and the speed limit of the lane. , so as to determine the speed deviation index of the vehicle to be corrected, so as to accurately know the degree of abnormality of the vehicle.

In one embodiment, as shown in FIG. 3 , it is a schematic flowchart of determining the corrected predicted driving trajectory of each vehicle to be corrected based on the environmental information of each corrected vehicle in one embodiment, including:

Step 302 , for each vehicle to be corrected, determine a set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected.

The environmental area information refers to the surrounding area information of the vehicle to be corrected. For example, the environmental area information refers to the information of the environmental area within the preset distance range of the vehicle to be corrected. For example, whether there is a sidewalk on the side of the vehicle to be corrected, or whether there is a non-driving area such as a flower bed on the side of the vehicle to be corrected.

Specifically, for each vehicle to be corrected, the computer device determines a set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected. The drivable area set includes a set of drivable areas within a preset distance. The drivable area refers to an area in which the vehicle to be corrected can travel in the future. For example, the right side of the lane where the vehicle to be corrected is located is a sidewalk, the left is other lanes, and there is no stationary obstruction in the lane where the vehicle to be corrected is located. Then, based on the environmental area information, it can be determined that the drivable areas of the vehicle to be corrected are the straight-traveling area and the left area.

Step 304: Determine the area probability value of each drivable area in the drivable area set based on the environmental vehicle information of the vehicle to be corrected.

Specifically, the computer device may determine the area probability value of each drivable area in the drivable area set based on the number of environmental vehicles of the vehicle to be corrected in each drivable area. Optionally, the computer device may determine the probability value of each drivable area based on the number of trajectory points of the vehicle in each drivable area of the vehicle to be corrected.

For example, the environmental vehicle information can be S _i =( _xi ,y _i ,θ _i ,vi ,a _i ,P' _i ), where x _i , y _i represent the position coordinates of the vehicle _i , and θ _i represents the position coordinates of the vehicle i Orientation, v _i represents the vector speed of vehicle i, a _i represents the vector acceleration of vehicle i, P' _i represents the predicted driving trajectory of vehicle i, P' _i ={x _p1 ,y _p1 ,x _p2 ,y _p2 ......x _pn , y _pn }, a total of n predicted trajectory points. And when the predicted travel trajectory of the vehicle i has been revised, P' _i represents the revised predicted travel trajectory. Then the total vehicle information S={S _e , S ₁ , S ₂ ...... S _m }, where Se is the vehicle information to be corrected, and m is the total number of vehicles in the environment.

Step 306: Determine the target driving area of the vehicle to be corrected based on the area probability value of each drivable area.

Specifically, the computer device may use the drivable area with the largest area probability value as the target driving area of the vehicle to be corrected. Or, according to the actual setting, the drivable area with the smallest area probability value may be used as the target driving area of the vehicle to be corrected. The target driving area refers to the area that the vehicle to be corrected will travel to in the future.

Step 308: Determine the corrected predicted driving trajectory of the vehicle to be corrected based on the target driving area.

Specifically, the computer device determines the revised predicted travel trajectory of the vehicle to be revised based on the target travel area. Some of the driving trajectory points in the revised predicted driving trajectory are located in the target driving area.

In the driving trajectory determination method in this embodiment, since not all the areas around the vehicle to be corrected are drivable areas, such as sidewalks, etc. are not the drivable areas of the vehicle to be corrected, then for each vehicle to be corrected, based on the information of the environment area Determine the set of drivable areas of the vehicle to be corrected, and obtain the area where the vehicle to be corrected can travel in the future; then determine the probability value of each drivable area according to the environmental vehicle information, so as to determine the target driving trajectory, and determine the to-be-corrected based on the target driving trajectory If the corrected predicted running trajectory of the vehicle is used, the obtained corrected predicted running trajectory is more accurate.

In one embodiment, the ambient vehicle information includes a predicted travel trajectory of the ambient vehicle. Determine the area probability value of each drivable area in the drivable area set based on the environmental vehicle information of the vehicle to be corrected, including:

In step (a1), a set of environmental vehicles of the vehicle to be corrected is obtained by searching within a preset distance range of the vehicle to be corrected, and the environmental vehicles in the set of environmental vehicles have a potential interaction relationship with the vehicle to be corrected.

The preset distance range may refer to a circular range surrounded by a radius with the center of the vehicle to be corrected as the origin and the preset distance as the origin. Or, take the center of the vehicle to be corrected as the origin, and the preset distance is a rectangular area surrounded by rectangular sides. The preset distance can be set according to actual needs. The specific preset distance may include the length of at least two vehicles, etc., but is not limited thereto.

Specifically, the computer obtains the set of environmental vehicles of the vehicle to be corrected by searching within a preset distance range of the vehicle to be corrected. The environmental vehicle in the environmental vehicle set has a potential interaction relationship with the vehicle to be corrected. The potential interaction relationship with the vehicle to be corrected means that there may be interaction with the vehicle to be corrected at a future time. For example, vehicle Y is included in the left lane of the vehicle to be corrected, and vehicle Z is included in the right lane of the vehicle to be corrected, then vehicle Y and vehicle Z are environmental vehicles of the vehicle to be corrected.

Step (a2), for each drivable area in the set of drivable areas, based on the predicted driving trajectory of the environmental vehicle in the set of environmental vehicles, determine the target trajectory point located in each drivable area.

The predicted driving trajectory includes at least two trajectory points.

Specifically, for each drivable area in the set of drivable areas, the computer device determines a target trajectory point located in each drivable area based on the predicted travel trajectories of the environmental vehicles in the set of environmental vehicles. For example, the drivable area set includes a drivable area A and a drivable area B. The environmental vehicle set includes environmental vehicle Y and environmental vehicle Z. Then, based on the predicted driving trajectory of the environmental vehicle Y, the trajectory point set I in the drivable area A is determined. Based on the predicted driving trajectory of the environmental vehicle Z, the trajectory point set II in the drivable area A is determined. Then the target trajectory points located in the drivable area A include the trajectory points in the trajectory point set I and the trajectory points in the trajectory point set II.

In step (a3), the area cost value of each drivable area is determined based on the target trajectory points located in each drivable area.

Among them, the area cost value is used to represent the cost that the vehicle to be corrected needs to pay for driving to the drivable area. A larger area cost value indicates that the drivable area should not be selected.

Specifically, the computer device determines the area cost value of each drivable area based on the number of target trajectory points located in each drivable area. The number of target trajectory points is positively correlated with the regional cost value. That is, when the number of target trajectory points is larger, the regional cost value is larger; when the number of target trajectory points is smaller, the regional cost value is smaller.

In step (a4), the area probability value corresponding to each drivable area is determined according to the area cost value of each drivable area.

Specifically, the computer device may determine the area probability value according to the ratio of the area cost value of each drivable area to the total area cost value corresponding to the set of drivable areas. For example, cost _left , cost _straight , and cost _right represent the regional cost values of the left lane change area, the straight area, and the right lane change area, respectively, and prob _left , prob _straight , and prob _right represent the left lane change area, the straight area, and the right lane change, respectively The area probability value for the area.

The driving trajectory determination method in this embodiment searches for a set of environmental vehicles that have a potential interactive relationship with the vehicle to be corrected within a preset distance range of the vehicle to be corrected, and determines the target trajectory points located in each drivable area, thereby determining The area cost value is obtained to obtain the cost paid by the vehicle to be corrected to travel to the drivable area, thereby determining the area probability value corresponding to each drivable area, thereby determining the target driving area and improving the accuracy of the predicted driving trajectory.

In one embodiment, determining the area cost value of each drivable area based on the target trajectory points located in each drivable area includes: obtaining a point weight corresponding to the target trajectory point; for a set of predicted trajectories corresponding to each environmental vehicle , obtain the trajectory probability value corresponding to each predicted trajectory; for each environmental vehicle, based on the sum of the point weights of the target trajectory points of each predicted trajectory and the trajectory probability value of each predicted trajectory of the environmental vehicle, obtain the environmental The vehicle driving cost value of the vehicle for the drivable area; the area cost value of each drivable area is determined based on the sum of the vehicle driving cost values of each environmental vehicle in the environmental vehicle set.

Among them, the point weight can be used to represent the number of trajectory points. For example, the point weight is 1. An environmental vehicle may correspond to at least one predicted trajectory to form a predicted trajectory set, and the trajectory probability values corresponding to each predicted driving trajectory of an environmental vehicle may be different. For example, the trajectory probability value corresponding to the predicted driving trajectory a of the environmental vehicle A is 0.6, and the trajectory probability value corresponding to the predicted driving trajectory b of the environmental vehicle A is 0.4.

Specifically, the computer device obtains the point weight corresponding to the target trajectory point. For the predicted trajectory set corresponding to each environmental vehicle, obtain the trajectory probability value corresponding to each predicted trajectory in the predicted trajectory set. For each environmental vehicle, the vehicle driving cost value of each environmental vehicle for the drivable area is obtained based on the summation of the point weights of the target trajectory points of each predicted driving trajectory and the trajectory probability value based on the predicted driving trajectory of the environmental vehicle. For example, the predicted travel trajectory set of the environmental vehicle A includes the predicted travel trajectory a and the predicted travel trajectory b. The trajectory probability value corresponding to the predicted driving trajectory a is 0.6, and the trajectory probability value corresponding to the predicted driving trajectory b is 0.4. The predicted driving trajectory a includes 3 target trajectory points located in the drivable area A, and the predicted driving trajectory b includes 4 target trajectory points located in the drivable area A, assuming that the point weight of each target trajectory point is 1. Then for the environmental vehicle A, the sum of the point weights based on the predicted driving trajectory a is 3, and the corresponding trajectory probability value is 0.6, and 3*0.6=1.8 is obtained. Based on the sum of the weights of the predicted driving trajectory b is 4, the corresponding trajectory probability value is 0.4, and 4*0.4=1.6 is obtained. Then, the vehicle driving cost of the environment vehicle A to the drivable area A is 1.6+1.8=3.4. Based on the sum of the vehicle travel cost values of the environmental vehicles in the environmental vehicle set, the computer device may determine the area cost value in response to the drivable area.

E.g,

Among them, w _car is the weight, M is the total number of vehicles in the environmental vehicle set, T is the total number of predicted driving trajectories, prob _t is the probability of the t-th predicted driving trajectory, N is the total number of predicted driving trajectories, p' _jti is the coordinate position of the i-th trajectory point of the t-th trajectory of vehicle j, and R is the driving area in the currently considered area. g(p' _jti , R) represents the point weight. g(p' _jti , R)=1 means that when the i-th point on the t-th trajectory of vehicle j is in the R area, the point weight of the point is 1, which means that the ith point on the t-th trajectory of vehicle j is 1. The i-th point is the target trajectory point, and the corresponding point weight is 1. And g(p' _jti , R)=0 means that when the i-th point on the t-th trajectory of vehicle j is not in the R area, the point weight corresponding to this point is 0, and the point is not the target trajectory point. p' _j represents the predicted travel trajectory of vehicle j. When the predicted travel trajectory of vehicle j has not been revised, the uncorrected predicted travel trajectory is used; when the predicted travel trajectory of vehicle j has been revised, the revised predicted travel trajectory is used. driving track. The vehicle to be corrected can detect and deduce the actual road traffic light situation in real time in the current environment, where R represents the drivable area mark and can pass the green light.

The driving trajectory determination method in this embodiment obtains a trajectory probability value corresponding to each predicted trajectory, corresponds to each environmental vehicle, and is based on the sum of the point weights of the target trajectory points of each predicted driving trajectory and each predicted trajectory of the environmental vehicle. The probability value of the driving trajectory is used to obtain the driving cost value of the environmental vehicle for the drivable area. In a word, the target trajectory point located in the drivable area is the trajectory point in the predicted driving trajectory of each environmental vehicle, so as to calculate the trajectory point located in each drivable area. The value corresponding to the target trajectory point in the area can be obtained to obtain the area cost value of the drivable area, which can more accurately calculate the cost of the vehicle traveling to the drivable area, and improve the accuracy of the predicted driving trajectory.

In one embodiment, determining the area cost value of each drivable area based on the sum of the vehicle driving cost values of the vehicles in each environment includes: for each drivable area, obtaining the lane change cost value of the vehicle to be corrected relative to the drivable area ; Obtain the lateral distance value between the center position of the vehicle to be corrected and the center line of the drivable area; determine the area cost value of the drivable area based on the lane change cost value, the lateral distance value and the sum of the vehicle travel cost values of the vehicles in each environment.

The lane change cost value is used to indicate whether the vehicle to be corrected can change lanes to the lane corresponding to the drivable area. For example, when the vehicle to be corrected can change lanes to the lane corresponding to the drivable area, the lane change cost value is 1; when the vehicle to be corrected cannot change lanes to the lane corresponding to the drivable area, the lane change cost value is 0.

Specifically, for each drivable area, the computer device obtains the lane-changing cost of the vehicle to be corrected relative to the drivable area. The computer device obtains the value of the lateral distance between the center position of the vehicle to be corrected and the center line of the drivable area. The computer device may convert the lateral distance value into a value representing the cost value. The computer device determines the area cost value of the drivable area based on the lane change cost value, the lateral distance value, and the vehicle travel cost value of each environmental vehicle. Then the sum of lane change cost value, lateral distance value and vehicle travel cost value is positively correlated with the regional cost value of the drivable area, respectively.

For example, the lane change cost C _change :

Where w _change is the weight, and kChangeLaneCost is a constant. That is, when the drivable area is the left or right lane changing area, the lane changing cost is kChangeLaneCost; when the drivable area is the straight area, there is no lane changing cost.

Lateral distance value C _lateral :

C _lateral =w _{lat_dis} *|lat_dis|

where w _{lat_dis} is the weight, and lat_dis is the lateral distance between the center position of the vehicle to be corrected and the center line of the drivable area.

The sum C _car of the vehicle travel cost of each environmental vehicle:

The area cost cost of the drivable area:

cost=C _change +C _lateral +C _car

In the driving trajectory determination method in this embodiment, for each drivable area, the lane change cost value of the vehicle to be corrected relative to the drivable area is obtained, and the lateral distance value between the center position of the vehicle to be corrected and the center line of the drivable area is obtained , the regional cost value of the drivable area can be determined based on the sum of the lane change cost value, the lateral distance value and the vehicle driving cost value of each environmental vehicle, which can more accurately calculate the regional cost value relative to a certain drivable area and improve the predicted driving. accuracy of the trajectory.

In one embodiment, the set of environmental vehicles of the vehicle to be corrected is obtained by searching within a preset distance range of the vehicle to be corrected, including:

The vehicle in the first area is used as the environmental vehicle of the vehicle to be corrected, and the set of environmental vehicles corresponding to the first area is obtained; The area where the vehicle is located is in the same area;

Taking the vehicle with the closest distance to the vehicle to be corrected in each second area as an environmental vehicle, and obtaining a set of environmental vehicles corresponding to the second area; the second area is an area other than the first area within the preset distance range;

The environmental vehicle set corresponding to the first area and the environmental vehicle set corresponding to the second area are used as the environmental vehicle set of the vehicle to be corrected.

The positive side of the vehicle to be corrected refers to the direction perpendicular to the lane of the vehicle to be corrected. Specifically, it may be the direction corresponding to the door of the vehicle to be corrected. And the positive side of the vehicle to be corrected refers to the direction on the same horizontal line as the area where the vehicle to be corrected is located. The preset distance range can be configured according to requirements.

Specifically, the computer device takes a vehicle in the first area as an environmental vehicle of the vehicle to be corrected, and obtains a set of environmental vehicles corresponding to the first area. The computer device takes the vehicle with the closest distance to the vehicle to be corrected in each second area as an environmental vehicle, and obtains a set of environmental vehicles corresponding to the second area. For example, the second area includes area M and area N. Then, taking the vehicle closest to the vehicle to be corrected in the area M as the environmental vehicle, and taking the vehicle closest to the vehicle to be corrected in the area N as the environmental vehicle, two environmental vehicles corresponding to the second area are obtained. The computer device takes the environmental vehicle set corresponding to the first area and the environmental vehicle set corresponding to the second area as the environmental vehicle set of the vehicle to be corrected.

In this embodiment, as shown in FIG. 4 , it is a schematic diagram of a preset distance range in an embodiment. With the vehicle to be corrected as the center, the left front area LF, the front right area CF, the right front area RF, the left area L, the center area C, the right area R, the left rear area LR, the right rear area CR, and the right rear area RR are respectively generated. The area size satisfies l1+l2+l3=forward_distance+backward_distance, and w1, w2, and w3 are the widths of each lane, respectively. The L, C, and R regions are the first regions. Both the L region and the R region are positive lateral regions. For regions L, C, and R, it is considered that all vehicles in these regions have a potential interaction relationship with the vehicle to be corrected, and they are added to the environmental vehicle set S _consider . LF, CF, RF, LR, CR, and RR are the second area. For the regions LF, CF, RF, LR, CR, and RR, it is considered that the vehicles closest to the target vehicle in each region have a potential interaction relationship with the target vehicle, and they are added to the set of environmental vehicles. If part of the car falls in the first area and the other part falls in the second area, then the car can be added to the first area; or the area of the car in each area is determined, and the area where the area is larger is used as the target area .

The driving trajectory determination method in this embodiment divides an area within a preset distance into a first area and a second area, wherein the first area is within the preset distance and on the right side of the vehicle to be corrected The area and the area in the same area as the area where the vehicle to be corrected is located, that is, there is a high possibility of potential interaction between the first area and the vehicle to be corrected, so the vehicle in the first area is used as the environment vehicle of the vehicle to be corrected; However, the relationship between the vehicles in the second area and the vehicle to be corrected is relatively weak. Therefore, the vehicle with the closest distance to the vehicle to be corrected in each second area is used as the environmental vehicle, thereby reducing the amount of calculation of the trajectory of the environmental vehicle and improving the trajectory. forecast efficiency.

In one embodiment, determining the set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected includes: taking the location of the vehicle to be corrected as a starting point, determining the through area and lane change area corresponding to the vehicle to be corrected, Areas and lane change areas are added to the set of drivable areas for the vehicle to be corrected.

Specifically, the straight-forward area refers to an area corresponding to the forward direction of the vehicle. Specifically, it may be located in the front area of the lane where the vehicle to be corrected is currently located. The lane change area refers to the front area next to the lane where the vehicle is located, which can be used for the vehicle to change lanes. The lane changing area may include a left lane changing area and a right lane changing area. In some cases, the vehicle may only have a left lane change area, or the vehicle may only have a right lane change area, or the vehicle may have no lane change area. For example, if one side of the vehicle is a sidewalk, or one side of the vehicle is in the middle of the road, the vehicle may only have one lane change area. When the road where the vehicle is located is a one-way street, there is no lane changing area for the vehicle.

For example, take the position of the vehicle to be corrected as the origin, extend forward_distance distance forward and backward_distance distance backward, as the straight travel region Region _straight . The left lane change area Region _left and the right lane change area Region _right are respectively generated on the left adjacent lane and the right adjacent lane with the same extension distance. If there is no left adjacent lane or right adjacent lane in the current lane, the corresponding Region will not be generated. Finally, the drivable area R={Region _straight , Region _left , Region _right } is obtained.

The driving trajectory determination method in this embodiment takes the position of the vehicle to be corrected as a starting point, determines the through area and lane change area corresponding to the vehicle to be corrected, and adds the through area and the lane change area to the set of drivable areas of the vehicle to be corrected , then the drivable area of the vehicle is determined, so that the revised predicted travel trajectory can be further predicted.

In one embodiment, determining the corrected predicted driving trajectory of the vehicle to be corrected based on the target driving area includes: acquiring the current location of the vehicle to be corrected; acquiring the future location of the vehicle to be corrected, where the future location is located in the target driving area ; Determine the corrected predicted travel trajectory of the vehicle to be corrected based on the current location and the future location.

The future location refers to the location at the future moment. The future location may specifically be located in the center of the target driving area.

Specifically, the current location can be represented by a traveling representation of coordinates. The computer device acquires the current location of the vehicle to be corrected through positioning, and acquires the future location of the vehicle to be corrected, where the future location is located in the target driving area. The computer device performs calculations based on the current location and the future location, and determines the corrected predicted travel trajectory of the vehicle to be corrected.

Optionally, based on the current time and the corresponding current location, the current driving speed, and the future time and the corresponding future location and future driving speed, the corrected predicted driving trajectory of the vehicle to be corrected is determined.

Optionally, obtain the current time of the vehicle to be corrected, the corresponding current location, the current driving speed, and the future time and the corresponding future location and future driving speed;

Inputting the position parameter value corresponding to the first direction in the current position and the current moment into the first preset function relational formula to obtain the first relational formula;

Inputting the first direction component of the current moment and the current driving speed into the first-order derivative function of the first preset functional relationship to obtain the second relationship;

Inputting the first direction component in the future time and the future driving speed into the first derivative function of the first preset functional relationship to obtain the fourth relationship;

Based on the first relational expression, the second relational expression and the fourth relational expression, a revised predicted driving trajectory in the first direction is obtained;

Inputting the position parameter value corresponding to the second direction in the current position and the current moment into the second preset function relational formula to obtain the sixth relational formula;

Inputting the second direction component of the current moment and the current driving speed into the first derivative function of the second preset functional relationship to obtain the seventh relationship;

Inputting the second direction component corresponding to the future time and the future position into the second preset functional relationship to obtain the ninth relationship;

Inputting the second direction component in the future time and the future driving speed into the first-order derivative function of the second preset functional relationship to obtain the tenth relationship;

Based on the sixth relational expression, the seventh relational expression, the ninth relational expression and the tenth relational expression, the revised predicted travel trajectory in the second direction is obtained.

The driving trajectory determination method in this embodiment obtains the current location and future location of the vehicle to be corrected, the future location is located in the target driving area, and the corrected predicted driving of the vehicle to be corrected is determined based on the current location and the future location. If the trajectory is selected, the predicted driving trajectory can be corrected based on the target driving area, and the accuracy of the predicted driving trajectory can be improved.

In one embodiment, acquiring the current location of the vehicle to be corrected includes: taking the center line of the target driving area as the coordinate axis in the first direction in the coordinate system, and the axis perpendicular to the first direction as the coordinate axis in the second direction , establish the vehicle traveling coordinate system; obtain the current location of the vehicle to be corrected under the vehicle traveling coordinate system. Acquiring the future location of the vehicle to be corrected includes: acquiring the future location of the vehicle to be corrected in the vehicle traveling coordinate system. Determining the corrected predicted travel trajectory of the vehicle to be corrected based on the current location and the future location includes: determining the corrected predicted travel trajectory of the vehicle to be corrected based on the current location and the future location in the vehicle travel coordinate system.

The center line of the target driving area refers to the center line of the lane corresponding to the target driving area. The coordinate axis in the first direction may be a longitudinal coordinate axis, and the coordinate axis in the second direction may be a horizontal coordinate axis. Alternatively, the coordinate axis in the first direction may be a horizontal coordinate axis, and the coordinate axis in the second direction may be a vertical coordinate axis.

Taking the trajectory generation of the left lane change area as an example to illustrate the generation method: the Frenet–Serret coordinate system is established based on the center line of the left lane change area. The direction along the center line is the s direction, and the direction perpendicular to the center line is the l direction. Then a trajectory can be expressed as pred=(s1,l1,s2,l2...sn,ln) in this coordinate system. The computer device acquires the current position of the vehicle to be corrected under the vehicle traveling coordinate system. The computer device acquires the future location of the vehicle to be corrected in the vehicle traveling coordinate system, and determines the corrected predicted traveling trajectory of the vehicle to be corrected based on the current location and the future location in the vehicle traveling coordinate system.

In the method for determining the driving trajectory in this embodiment, since the lane is curved in the world coordinate system, the calculation based on various offsets is more complicated. Therefore, the center line of the target driving area is used as the first coordinate system in the coordinate system. The coordinate axis of the direction, the axis perpendicular to the first direction is used as the coordinate axis of the second direction, the vehicle driving coordinate system is established, and the vehicle to be corrected and the current position and future position in the vehicle driving coordinate system are determined. Based on the vehicle driving The coordinate system is used for calculation, which can reduce the calculation amount of the predicted driving trajectory.

In one embodiment, the preset functional relationship includes a first preset functional relationship and a second preset functional relationship, and the first preset functional relationship includes a time parameter and a position parameter in the first direction; the second function The relational expression includes a time parameter and a position parameter in the second direction.

Determine the corrected predicted driving trajectory of the vehicle to be corrected based on the current location and future location, including:

Obtain the current time of the vehicle to be corrected, the corresponding current driving speed, the current driving acceleration, and the future time and the corresponding future driving speed and future driving acceleration;

Inputting the first direction component of the current moment and the current driving speed into the first derivative function of the first preset functional relationship to obtain the second relationship;

Inputting the first direction component of the current moment and the current driving acceleration into the second-order derivative function of the first preset functional relationship to obtain a third relationship;

Inputting the first direction component in the future time and the future driving acceleration into the second-order derivative function of the first preset functional relationship to obtain a fifth relationship;

Based on the first relational expression, the second relational expression, the third relational expression, the fourth relational expression and the fifth relational expression, obtain the revised predicted driving trajectory in the first direction;

Inputting the second direction component of the current moment and the current driving acceleration into the second-order derivative function of the second preset functional relationship to obtain the eighth relationship;

Inputting the second direction component in the future time and the future driving acceleration into the second-order derivative function of the second preset functional relationship to obtain the eleventh relationship;

Based on the sixth relational expression, the seventh relational expression, the eighth relational expression, the ninth relational expression, the tenth relational expression and the eleventh relational expression, a revised predicted driving trajectory in the second direction is obtained;

The modified predicted travel trajectory of the vehicle to be corrected is obtained based on the modified predicted travel trajectory in the first direction and the modified predicted travel trajectory in the second direction.

Wherein, the first preset functional relationship refers to a functional relationship between a position parameter and a time parameter in the direction of the center line of the target driving area. Specifically, it may refer to a functional relationship in the direction in which the vehicle is traveling. The first preset functional relationship is used to represent the position of the vehicle to be corrected in the direction of travel of the vehicle at time t. The second preset functional relationship is a functional relationship between the position parameter and the time parameter in the direction perpendicular to the center line of the target travel area. Specifically, it may refer to a functional relationship that is perpendicular to the driving direction of the vehicle. The second preset functional relationship is used to indicate which position of the vehicle to be corrected is perpendicular to the traveling direction of the vehicle at time t. The corrected predicted travel trajectory in the first direction and the corrected predicted travel trajectory in the second direction are the corrected predicted travel trajectory of the vehicle to be corrected.

Specifically, an example is used for description, and the Frenet-Serret coordinate system is established based on the center line of the left lane changing area. The direction along the center line is the s direction, and the direction perpendicular to the center line is the l direction. Then a trajectory can be expressed as pred=(s1,l1,s2,l2...sn,ln) in this coordinate system,

Where s=f(t), l=g(t), t is the time. Thus the modified predicted travel trajectory in the first direction can be denoted by f(t) and the modified predicted travel trajectory in the second direction can be denoted by g(t).

make

f(t)=a*t ⁴ +b*t ³ +ct ² +dt+e

Let time 0 be the current time, time 5 seconds be the future time, s ₀ , ds ₀ , and dds ₀ are the position, speed, and acceleration of the vehicle in the s direction at time 0, respectively, and ds ₅ , dds ₅ are the time 5 seconds The vehicle is in the s direction speed and acceleration. available

The first relational expression: f(0)=s ₀

The second relational formula: f(0)'=ds ₀

The third relational formula: f(0)”=dds ₀

The fourth relational formula: f(5)'=ds ₅

The fifth relational formula: f(5)”=dds ₅

Then the current driving speed and the current driving acceleration can be directly known. In 5 seconds, the driving of the vehicle to be corrected can be regarded as driving at a constant speed, so the value of ds ₅ is set to the value of ds ₀ . The value of dds ₅ can be set to 0. From this, the values of a, b, c, d, and e in f(t) can be obtained, so as to obtain the revised predicted driving trajectory in the first direction.

The same order

g(t)=a*t ⁵ +b*t ⁴ +ct ³ +dt ² +et+f

Let l ₀ , dl ₀ , ddl ₀ be the position, velocity, and acceleration of the vehicle in the l direction at time 0, and l ₅ , dl ₅ , and ddl ₅ be the position, velocity, and acceleration of the vehicle in the l direction at 5 seconds, then there are:

The sixth relational formula: g(0)=l ₀

The seventh relational formula: g(0)'=dl ₀

The eighth relational formula: g(0)”=ddl ₀

The ninth relational formula: g(5)=l ₅

Tenth relational formula: g(5)'=dl ₅

Eleventh relational formula: g(5)”=ddl ₅

In the second preset functional relationship, it is assumed that the vehicle to be corrected reaches the center of the target area within 5 seconds, then there are l ₅ =0, dl ₅ =0, ddl ₅ =0, that is, the future position is the position of the origin, The future lateral velocity component is 0, and the future lateral acceleration component is also 0. So solving the system of equations gives g(t).

After obtaining pred=(s1,l1,s2,l2...sn,ln) through f(t) and g(t), the sl coordinate can be obtained based on the center line of the lane in the left lane change area as the xy coordinate Revised Predicted Trajectory

In the method for determining the driving trajectory in this embodiment, two directions are determined based on the current time, current location, current driving speed, current driving acceleration, and parameter values of future time, future location, future driving speed, and future driving acceleration. The predicted driving trajectory above can be obtained to obtain a more accurate corrected predicted driving trajectory.

In one embodiment, the driving trajectory determination method further includes: when the number of drivable areas in the drivable area set is one, taking the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected as the vehicle to be corrected the actual driving trajectory.

Specifically, when the number of drivable areas in the set of drivable areas is one, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected is directly used as the actual driving trajectory of the vehicle to be corrected. The actual driving trajectory is used for automatic driving or instructing the driver to drive.

In the driving trajectory determination method in this embodiment, when the number of drivable areas in the drivable area set is one, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected is directly used as the actual driving trajectory of the vehicle to be corrected. If the driving trajectory is determined, the target driving area does not need to be determined, and the driving trajectory correction of the vehicle to be corrected is not required, thereby improving the driving trajectory determination efficiency.

In one embodiment, the method for obtaining the set of vehicles to be corrected includes: taking the target autonomous driving vehicle as a reference, determining a preset distance range corresponding to the target autonomous driving vehicle; adding vehicles within the preset distance range to the vehicle to be corrected Fixed vehicle collection. The driving trajectory determination method further includes: correcting the predicted driving trajectory of the target automatic driving vehicle based on the corrected predicted driving trajectory of each vehicle to be corrected in the set of vehicles to be corrected.

Specifically, the computer device is located in the target autonomous vehicle. The target autonomous driving vehicle can be used as a reference, and the computer device determines a preset distance range corresponding to the target autonomous driving vehicle, and uses the vehicles within the preset distance range as the set of vehicles to be corrected. The computer device adds vehicles within 10 meters of the target autonomous vehicle to the set of vehicles to be corrected corresponding to the target autonomous vehicle. The computer device corrects the predicted travel trajectory of the target autonomous vehicle based on the corrected predicted travel trajectory of each to-be-corrected vehicle in the to-be-corrected vehicle set. For example, with the center of gravity of the target autonomous vehicle as the center and the orientation of the autonomous vehicle as the orientation, a rectangular area Rec with length and width of L and H is established, and the vehicles in the area are used as the vehicle set V _Rec to be corrected.

The driving trajectory determination method in this embodiment uses the automatic driving vehicle as a reference to determine a set of vehicles to be corrected located within a preset distance range. Since the predicted driving trajectory of the vehicle to be corrected has been corrected, the predicted driving trajectory of the target autonomous driving vehicle is It should also be corrected to improve the accuracy of the predicted trajectory of the target autonomous vehicle.

In one embodiment, the driving trajectory determination method further includes: acquiring a human driving trajectory, comparing the revised predicted driving trajectory with the human driving trajectory, and obtaining a comparison result; when the comparison result is within a preset gap range, determining the The corrected predicted travel trajectory is a qualified trajectory; when the comparison result is not within the preset gap range, the corrected predicted travel trajectory is determined to be an unqualified trajectory, and the corrected predicted travel trajectory is reported. In this embodiment, the revised predicted driving trajectory can be evaluated by comparing the human driving trajectory with the corrected predicted driving trajectory. If it is within the preset gap range, it means that it is a good trajectory. , it is explained that the algorithm for predicting the driving trajectory needs to be redesigned.

In one embodiment, as shown in FIG. 5 , it is a schematic flowchart of a method for determining a driving trajectory in another embodiment, which includes:

Step 502 , taking the target automatic driving vehicle as a reference, obtain the environmental vehicle information.

Step 504: Obtain a set of vehicles to be corrected corresponding to the target autonomous vehicle.

Step 506: Calculate the abnormal state index of each vehicle to be corrected in the set of vehicles to be corrected, and arrange them according to the degree of abnormality from small to large.

Step 508, is the set of vehicles to be corrected empty?

Step 510, when the set of vehicles to be corrected is not empty, obtain the first vehicle in the set of vehicles to be corrected, determine the corrected predicted travel trajectory according to the environmental information, move the vehicle to be corrected out of the set of vehicles to be corrected, and return to the execution step 508.

Step 512 , when the set of vehicles to be corrected is empty, output the corrected predicted travel trajectories of the vehicles to be corrected.

In this embodiment, predicting the future trajectory of the surrounding environment is a big problem, especially in highly dynamic and interactive scenarios, the behavior and trajectory prediction problem has not been well solved, even if it is perfect Decision-making, planning, and control are unlikely to be safe and efficient in practical applications, so in this embodiment of the application, the trajectory of the autonomous vehicle will be combined to correct the predicted trajectory of the current input surrounding predicted objects to ensure a safe and reliable decision-making plan. . By calculating the abnormal state index of environmental vehicles, an abnormal state transfer model between vehicles is established to describe the interaction between vehicles. And the prediction result of the environmental vehicle is corrected according to the current vehicle interaction relationship. Through the solution in the embodiment of the present application, the interaction relationship of environmental vehicles under complex road conditions can be quickly established based on the abnormal state transfer model, avoiding the problem that the source of the change of the traffic flow state cannot be determined due to the coupling of the interaction relationship of each vehicle. Combined with the vehicle interaction relationship and the vehicle prediction information based on the motion model, more accurate prediction results can be obtained, thereby improving the input information quality and output instruction quality of the decision planning module. The correction starts from vehicles with a small degree of abnormality. One is to reflect the impact of abnormal conditions in the predicted traffic flow on normal vehicles. Secondly, the corrected trajectory of vehicles with a small degree of abnormality has almost no effect on vehicles with a large degree of abnormality. Large vehicles will affect the trajectory of the entire traffic flow after correction, so the vehicle with small abnormality is corrected first, and the calculation amount can be reduced in the calculation of predicted driving trajectory; and in trajectory prediction, vehicles with small abnormality are easier to leave. Because of the abnormal site, it is necessary to perform trajectory correction from vehicles with a small degree of abnormality. This scheme models the abnormality index of surrounding vehicles, and establishes the interaction relationship between surrounding vehicles according to the abnormality index, so that the interaction relationship between vehicles can be found relatively quickly; some surrounding vehicles are selected for trajectory correction to save computing resources; the environmental semantic map is used to improve the trajectory Correction accuracy; select the matching dynamic model according to the perceived vehicle type to verify the feasibility of the correction trajectory; establish the algorithm closed-loop verification evaluation standard to make the results more realistic.

In one embodiment, as shown in FIG. 6 , it is a schematic flowchart of determining a revised predicted driving trajectory based on environmental information in one embodiment, which includes:

Step 602, for each vehicle to be corrected, input the environmental information of the vehicle to be corrected.

Step 604 , determining a set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected.

Step 606, is the number of drivable areas in the drivable area set equal to 1?

Step 608 , when the number of drivable areas in the drivable area set is 1, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected is taken as the actual driving trajectory of the vehicle to be corrected.

Step 610 , when the number of drivable areas in the drivable area set is not 1, search for an environmental vehicle set that has a potential interaction relationship with the vehicle to be corrected.

Step 612: Calculate the driving cost value of each drivable area based on the environmental vehicle information, and determine the area probability value based on the driving cost value.

Step 614: Determine the target area according to the area probability value, and fit the revised predicted driving trajectory corresponding to the target area.

In this embodiment, a set of environmental vehicles with a potential interactive relationship with the vehicle to be corrected is searched within the preset distance range of the vehicle to be corrected, and the target trajectory points located in each drivable area are determined to determine the area cost value, and obtain The cost paid by the vehicle to be corrected for driving to the drivable area is determined, thereby determining the area probability value corresponding to each drivable area, thereby determining the target driving area and improving the accuracy of the predicted driving trajectory.

In one embodiment, a driving trajectory determination method includes:

Step (b1), with the target autonomous driving vehicle as a reference, determine a preset distance range corresponding to the target autonomous driving vehicle.

Step (b2), adding vehicles within a preset distance range to the set of vehicles to be corrected.

In step (b3), vehicle travel information of each vehicle to be corrected in the vehicle set to be corrected is acquired.

In step (b4), the lateral distance deviation index of each vehicle to be corrected is determined based on the lateral distance of each vehicle to be corrected from the center line of the lane where it is located.

In step (b5), the lateral speed deviation index of each vehicle to be corrected is determined based on the current vehicle lateral speed of each vehicle to be corrected.

In step (b6), the lane speed limit corresponding to the lane where each vehicle to be corrected is located is obtained.

In step (b7), the speed deviation value of each vehicle to be corrected is determined based on the current traveling speed and the speed limit of the lane.

Step (b8), determining the running speed deviation index of each vehicle to be corrected based on the speed deviation value.

In step (b9), an orientation deviation index of each vehicle to be corrected is determined based on the current vehicle orientation of each vehicle to be corrected and the orientation of the centerline of the lane where the vehicle to be corrected is located.

Step (b10): Determine the abnormal state index of each vehicle to be corrected based on the lateral distance deviation index, lateral speed deviation index, driving speed deviation index and orientation deviation index of each vehicle to be corrected.

In step (b11), the environmental information of each vehicle to be corrected in the set of vehicles to be corrected is acquired.

Step (b12), according to the order of the abnormal state index in the abnormal state index sequence, for each vehicle to be corrected, and the position of the vehicle to be corrected as the starting point, determine the through area and lane change area corresponding to the vehicle to be corrected, and the lane-changing area are added to the set of drivable areas of the vehicle to be corrected, and the abnormal state index sequence is arranged in order from weak to strong according to the abnormal degree represented by each abnormal state index.

In step (b13), the vehicle in the first area is used as the environmental vehicle of the vehicle to be corrected, and the set of environmental vehicles corresponding to the first area is obtained, and the first area is within the preset distance range and on the side of the vehicle to be corrected. area and areas in the same area as the area where the vehicle to be corrected is located.

In step (b14), the vehicle with the closest distance to the vehicle to be corrected in each second area is used as an environmental vehicle, and a set of environmental vehicles corresponding to the second area is obtained, and the second area is within the preset distance range except the first area. area.

Step (b15), take the environmental vehicle set corresponding to the first area and the environmental vehicle set corresponding to the second area as the environmental vehicle set of the vehicle to be corrected; the environmental vehicles in the environmental vehicle set have a potential interaction relationship with the vehicle to be corrected.

In step (b16), for each drivable area in the drivable area set, based on the predicted travel trajectories of the environmental vehicles in the environmental vehicle set, determine the target trajectory points located in each drivable area.

In step (b17), the point weight corresponding to the target trajectory point is obtained.

Step (b18), for the predicted trajectory set corresponding to each environmental vehicle, obtain the trajectory probability value corresponding to each predicted trajectory.

Step (b19), for each environmental vehicle, based on the sum of the point weights of the target trajectory points of each predicted driving trajectory and the trajectory probability value of each predicted driving trajectory of the environmental vehicle, obtain the driving of the environmental vehicle for the vehicle in the drivable area. cost value.

Step (b20), for each drivable area, obtain the lane change cost value of the vehicle to be corrected relative to the drivable area.

Step (b21), acquiring the value of the lateral distance between the center position of the vehicle to be corrected and the center line of the drivable area.

In step (b22), the area cost value of the drivable area is determined based on the lane change cost value, the lateral distance value and the sum of the vehicle travel cost values of each environmental vehicle.

In step (b23), the area probability value corresponding to each drivable area is determined according to the area cost value of each drivable area.

In step (b24), the target driving area of the vehicle to be corrected is determined based on the area probability value of each drivable area.

In step (b25), the center line of the target driving area is used as the coordinate axis of the first direction in the coordinate system, and the axis perpendicular to the first direction is used as the coordinate axis of the second direction, and the vehicle driving coordinate system is established.

In step (b26), the current location of the vehicle to be corrected in the vehicle traveling coordinate system is acquired.

In step (b27), the future location of the vehicle to be corrected in the vehicle driving coordinate system is acquired, and the future location is located in the target driving area.

Step (b28), in the vehicle travel coordinate system, obtain the current time of the vehicle to be corrected, the corresponding current travel speed, the current travel acceleration, and the future time and the corresponding future travel speed and future travel acceleration.

In step (b29), the position parameter value corresponding to the first direction in the current position and the current time are input into the first preset function relational expression to obtain the first relational expression.

Step (b30), inputting the first direction component of the current moment and the current driving speed into the first derivative function of the first preset functional relational expression to obtain the second relational expression.

Step (b31), inputting the current moment and the first direction component of the current driving acceleration into the second-order derivative function of the first preset functional relational expression to obtain a third relational expression.

Step (b32), inputting the first direction component in the future time and the future driving speed into the first derivative function of the first preset functional relational expression to obtain a fourth relational expression.

Step (b33), inputting the first direction component in the future time and the future driving acceleration into the second-order derivative function of the first preset functional relational expression to obtain a fifth relational expression.

Step (b34), based on the first relational expression, the second relational expression, the third relational expression, the fourth relational expression and the fifth relational expression, obtain a revised predicted travel trajectory in the first direction.

In step (b35), the position parameter value corresponding to the second direction in the current position and the current time are input into the second preset function relational expression to obtain a sixth relational expression.

Step (b36), inputting the second direction component of the current moment and the current driving speed into the first derivative function of the second preset functional relationship to obtain a seventh relationship.

Step (b37), inputting the second direction component of the current moment and the current driving acceleration into the second-order derivative function of the second preset functional relationship to obtain an eighth relationship.

In step (b38), the second direction component corresponding to the future time and the future position is input into the second preset functional relational expression to obtain a ninth relational expression.

Step (b39), inputting the second direction component in the future time and the future driving speed into the first derivative function of the second preset functional relational expression to obtain the tenth relational expression.

Step (b40), inputting the second direction component in the future time and the future driving acceleration into the second-order derivative function of the second preset functional relationship to obtain the eleventh relationship.

Step (b41), based on the sixth relational expression, the seventh relational expression, the eighth relational expression, the ninth relational expression, the tenth relational expression and the eleventh relational expression, obtain a revised predicted driving trajectory in the second direction ;

In step (b42), a modified predicted traveling trajectory of the vehicle to be corrected is obtained based on the modified predicted traveling trajectory in the first direction and the modified predicted traveling trajectory in the second direction.

In step (b43), when the number of drivable areas in the drivable area set is one, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected is taken as the actual driving trajectory of the vehicle to be corrected.

Step (b44), correcting the predicted running trajectory of the target automatic driving vehicle based on the corrected predicted running trajectory of each vehicle to be corrected in the set of vehicles to be corrected.

It should be understood that although the above steps are displayed in sequence according to the numerical indication, these steps are not necessarily performed in sequence according to the numerical indication. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders.

In the method for determining the driving trajectory in this embodiment, in the normal driving traffic flow, vehicles with a high degree of abnormality are generally the source of changes in the traffic flow. For example, a stationary vehicle in front will cause the vehicle behind it to slow down and change lanes. The abnormal state It will be gradually transmitted from the source over time to affect the entire traffic flow, that is, vehicles with low abnormality will be affected by vehicles with high abnormality, while vehicles with high abnormality will be less affected by vehicles with low abnormality. Based on this principle, the embodiment determines the abnormal state index of each vehicle to be corrected based on the vehicle driving information of the vehicle to be corrected, and corrects the predicted trajectory from the vehicle with a low degree of abnormality based on the environmental information of each vehicle to be corrected, so as to predict the abnormality in the traffic flow in advance. The influence of the situation on the normal running vehicle is determined, so as to determine the corrected predicted running trajectory of the vehicle in the vehicle set to be corrected, and improve the accuracy of the predicted running trajectory of the vehicle.

It should be understood that although the steps in the flowcharts of FIGS. 2, 3, 5 and 6 are shown in sequence according to the arrows, these steps are not necessarily executed in the sequence shown by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and the steps may be executed in other orders. Moreover, at least a part of the steps in FIGS. 2, 3, 5 and 6 may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but may be executed at different times. These steps Alternatively, the order of execution of the stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the steps or stages in the other steps.

In one embodiment, as shown in FIG. 7, a driving trajectory determination device is provided, including: a first acquisition module 702, an abnormal state index determination module 704, a second acquisition module 706, and a correction module 708, wherein:

a first obtaining module 702, configured to obtain vehicle driving information of each vehicle to be corrected in the set of vehicles to be corrected;

an abnormal state index determination module 704, configured to determine the abnormal state index of each vehicle to be corrected based on the vehicle driving information of each vehicle to be corrected;

A second obtaining module 706, configured to obtain the environmental information of each vehicle to be corrected in the set of vehicles to be corrected;

The correction module 708 is configured to determine, according to the order of the abnormal state indexes in the abnormal state index sequence, the corrected predicted driving trajectory of each vehicle to be corrected based on the environmental information of each vehicle to be corrected; The abnormality degree represented by the abnormal state index is arranged from weak to strong.

In the driving trajectory determination device in this embodiment, in the normal driving traffic flow, vehicles with a high degree of abnormality are generally the source of changes in the traffic flow. For example, a stationary vehicle in front will cause the vehicle behind it to decelerate and change lanes. The abnormal state It will be gradually transmitted from the source over time to affect the entire traffic flow, that is, vehicles with low abnormality will be affected by vehicles with high abnormality, while vehicles with high abnormality will be less affected by vehicles with low abnormality. Based on this principle, the embodiment determines the abnormal state index of each vehicle to be corrected based on the vehicle driving information of the vehicle to be corrected, and corrects the predicted trajectory from the vehicle with a low degree of abnormality based on the environmental information of each vehicle to be corrected, so as to predict the abnormality in the traffic flow in advance. The influence of the situation on the normal running vehicle is determined, so as to determine the corrected predicted running trajectory of the vehicle in the vehicle set to be corrected, and improve the accuracy of the predicted running trajectory of the vehicle.

In one embodiment, the vehicle driving information includes the lateral distance from the center line of the lane, the current lateral speed of the vehicle, the current driving speed and the current vehicle orientation. The abnormal state index determination module 704 is configured to determine the lateral distance deviation index of each to-be-corrected vehicle based on the lateral distance of each to-be-corrected vehicle from the centerline of the lane where it is located; Speed deviation index; determine the speed deviation index of each vehicle to be corrected based on the current driving speed of each vehicle to be corrected; determine the orientation of each vehicle to be corrected based on the current vehicle orientation of each vehicle to be corrected and the centerline orientation of the lane where the vehicle to be corrected is located Deviation index; the abnormal state index of each vehicle to be corrected is determined based on the lateral distance deviation index, lateral speed deviation index, driving speed deviation index and orientation deviation index of each vehicle to be corrected.

In the device for determining the driving trajectory in this embodiment, in the normal driving state, the vehicle generally drives along the lane line and complies with the traffic regulations; since the lateral distance from the center line of the vehicle, the current lateral speed of the vehicle and the current vehicle orientation are all It can be used to characterize whether the vehicle has a tendency to change lanes, whether it will lose control, etc. Therefore, the lateral distance from the center line of the vehicle can be considered as part of the abnormal state, and the lateral distance deviation index, lateral speed deviation index and orientation deviation index can be obtained; current; The driving speed can be used to characterize whether the vehicle is driving faster or slower, etc. Therefore, the current driving speed is also used to characterize whether the vehicle is driving too fast or too slow, etc., so as to accurately determine the abnormal degree of the vehicle, so that the vehicle's abnormality can be determined in sequence. The revised predicted driving trajectory.

In one embodiment, the abnormal state index determination module 704 is configured to obtain the lane speed limit corresponding to the lane where each vehicle to be corrected is located; determine the speed deviation value of each vehicle to be corrected based on the current driving speed and the lane speed limit; based on the speed deviation The value determines the travel speed deviation index of each vehicle to be corrected.

In the driving trajectory determination device in this embodiment, the vehicle is driving on the lane. Generally, there are different speed limits in different road sections. Therefore, the speed deviation value of each vehicle to be corrected can be determined based on the current driving speed of the vehicle and the speed limit of the lane. , so as to determine the deviation index of the driving speed of the vehicle to be corrected, so as to accurately know the degree of abnormality of the vehicle.

In one embodiment, the correction module 708 includes a drivable area set determination unit, an area probability value determination unit, a target driving area determination unit and a trajectory determination unit, wherein the drivable area set determination unit is used for each vehicle to be corrected based on The environmental area information of the vehicle to be corrected determines the drivable area set of the vehicle to be corrected; the area probability value determination unit is used to determine the area probability value of each drivable area in the drivable area set based on the environmental vehicle information of the vehicle to be corrected; the target driving area The determining unit is used for determining the target driving area of the vehicle to be corrected based on the area probability value of each drivable area; the trajectory determining unit is used for determining the corrected predicted driving trajectory of the vehicle to be corrected based on the target driving area.

In the driving trajectory determination device in this embodiment, since the areas around the vehicle to be corrected are not all drivable areas, such as sidewalks and the like are not the drivable areas of the vehicle to be corrected, then for each vehicle to be corrected, based on the information of the environment area Determine the set of drivable areas of the vehicle to be corrected, and obtain the area where the vehicle to be corrected can travel in the future; then determine the probability value of each drivable area according to the environmental vehicle information, so as to determine the target driving trajectory, and determine the to-be-corrected based on the target driving trajectory If the corrected predicted running trajectory of the vehicle is used, the obtained corrected predicted running trajectory is more accurate.

In one embodiment, the ambient vehicle information includes a predicted travel trajectory of the ambient vehicle. The area probability value determination unit is used to search the set of environmental vehicles of the vehicle to be corrected within the preset distance range of the vehicle to be corrected, and the environmental vehicles in the set of environmental vehicles have a potential interaction relationship with the vehicle to be corrected; For each drivable area, determine the target trajectory points located in each drivable area based on the predicted driving trajectories of the environmental vehicles in the environmental vehicle set; determine the regional cost value of each drivable area based on the target trajectory points located in each drivable area ; Determine the area probability value corresponding to each drivable area according to the area cost value of each drivable area.

The driving trajectory determination device in this embodiment searches for a set of environmental vehicles that have a potential interactive relationship with the vehicle to be corrected within the preset distance range of the vehicle to be corrected, and determines the target trajectory points located in each drivable area, thereby determining The area cost value is obtained to obtain the cost paid by the vehicle to be corrected to travel to the drivable area, thereby determining the area probability value corresponding to each drivable area, thereby determining the target driving area and improving the accuracy of the predicted driving trajectory.

In one embodiment, the area probability value determination unit is used to obtain the point weight corresponding to the target track point; for the predicted track set corresponding to each environmental vehicle, obtain the track probability value corresponding to each predicted track; for each environment Vehicle, based on the sum of the point weights of the target trajectory points of each predicted driving trajectory and the trajectory probability value of each predicted driving trajectory of the environmental vehicle, the vehicle driving cost value of the environmental vehicle for the drivable area is obtained; The sum of the vehicle travel cost values of the environmental vehicles determines the area cost value of each drivable area.

The driving trajectory determination device in this embodiment obtains the trajectory probability value corresponding to each predicted trajectory, corresponds to each environmental vehicle, and is based on the sum of the point weights of the target trajectory points of each predicted driving trajectory and each predicted trajectory of the environmental vehicle. The probability value of the driving trajectory is used to obtain the driving cost value of the environmental vehicle for the drivable area. In a word, the target trajectory point located in the drivable area is the trajectory point in the predicted driving trajectory of each environmental vehicle, so as to calculate the trajectory point located in each drivable area. The value corresponding to the target trajectory point in the area is obtained to obtain the area cost value of the drivable area, which can more accurately calculate the cost of the vehicle traveling to the drivable area, and improve the accuracy of the predicted driving trajectory.

In one embodiment, the area probability value determination unit is configured to, for each drivable area, obtain the lane change cost value of the vehicle to be corrected relative to the drivable area; obtain the lateral direction of the center position of the vehicle to be corrected from the center line of the drivable area Distance value; the area cost value of the drivable area is determined based on the lane change cost value, the lateral distance value, and the sum of the vehicle travel cost values of each environmental vehicle.

The driving trajectory determination device in this embodiment, for each drivable area, obtains the lane change cost value of the vehicle to be corrected relative to the drivable area, and obtains the value of the lateral distance between the center position of the vehicle to be corrected and the center line of the drivable area , the regional cost value of the drivable area can be determined based on the sum of the lane change cost value, the lateral distance value and the vehicle driving cost value of each environmental vehicle, which can more accurately calculate the regional cost value relative to a certain drivable area and improve the predicted driving. accuracy of the trajectory.

In one embodiment, the area probability value determination unit is configured to use a vehicle in the first area as an environmental vehicle of the vehicle to be corrected, and obtain a set of environmental vehicles corresponding to the first area; the first area is within a preset distance range and In the area directly to the side of the vehicle to be corrected and the area in the same area as the area where the vehicle to be corrected is located; the vehicle with the closest distance to the vehicle to be corrected in each second area is used as an environmental vehicle to obtain a set of environmental vehicles corresponding to the second area; The second area is an area other than the first area within the preset distance range; the environmental vehicle set corresponding to the first area and the environmental vehicle set corresponding to the second area are used as the environmental vehicle set of the vehicle to be corrected.

The driving trajectory determination device in this embodiment divides an area within a preset distance into a first area and a second area, wherein the first area is within the preset distance and on the right side of the vehicle to be corrected The area and the area in the same area as the area where the vehicle to be corrected is located, that is, there is a high possibility of potential interaction between the first area and the vehicle to be corrected, so the vehicle in the first area is used as the environment vehicle of the vehicle to be corrected; However, the relationship between the vehicles in the second area and the vehicle to be corrected is relatively weak. Therefore, the vehicle with the closest distance to the vehicle to be corrected in each second area is used as the environmental vehicle, thereby reducing the amount of calculation of the trajectory of the environmental vehicle and improving the trajectory. forecast efficiency.

In one embodiment, the drivable area determination unit is configured to take the position of the vehicle to be corrected as a starting point, determine the through area and the lane change area corresponding to the vehicle to be corrected, and add the through area and the lane change area to the drivable area of the vehicle to be corrected in the regional collection.

The driving trajectory determination device in this embodiment takes the position of the vehicle to be corrected as a starting point, determines the through area and lane change area corresponding to the vehicle to be corrected, and adds the through area and the lane change area to the set of drivable areas of the vehicle to be corrected , then the drivable area of the vehicle is determined, so that the revised predicted travel trajectory can be further predicted.

In one embodiment, the correction module 708 is configured to acquire the current location of the vehicle to be corrected; acquire the future location of the vehicle to be corrected, where the future location is located in the target driving area; determine the vehicle to be corrected based on the current location and the future location The revised predicted driving trajectory.

The driving trajectory determination device in this embodiment acquires the current location and future location of the vehicle to be corrected, the future location is located in the target driving area, and determines the corrected predicted driving of the vehicle to be corrected based on the current location and the future location If the trajectory is selected, the predicted driving trajectory can be corrected based on the target driving area, and the accuracy of the predicted driving trajectory can be improved.

In one embodiment, the correction module 708 is configured to use the center line of the target driving area as the coordinate axis of the first direction in the coordinate system, and the axis perpendicular to the first direction as the coordinate axis of the second direction to establish the vehicle driving coordinate system ; Obtain the current position of the vehicle to be corrected in the vehicle traveling coordinate system; obtain the future position of the vehicle to be corrected in the vehicle traveling coordinate system; in the vehicle traveling coordinate system, determine the to-be-corrected based on the current position and future position The revised predicted driving trajectory of the vehicle.

In the driving trajectory determination device in this embodiment, since the lane is curved in the world coordinate system, the calculation based on various offsets is more complicated. Therefore, the center line of the target driving area is used as the first coordinate system in the coordinate system. The coordinate axis of the direction, the axis perpendicular to the first direction is used as the coordinate axis of the second direction, the vehicle driving coordinate system is established, and the vehicle to be corrected and the current position and future position in the vehicle driving coordinate system are determined. Based on the vehicle driving The coordinate system is used for calculation, which can reduce the calculation amount of the predicted driving trajectory.

In one embodiment, the preset functional relationship includes a first preset functional relationship and a second preset functional relationship, and the first preset functional relationship includes a time parameter and a position parameter in the first direction; the second function The relational expression includes a time parameter and a position parameter in the second direction. The correction module 708 is used to obtain the current time of the vehicle to be corrected and the corresponding current driving speed, the current driving acceleration, and the future time and the corresponding future driving speed and future driving acceleration;

The driving trajectory determination device in this embodiment determines and obtains two directions based on the current time, current location, current driving speed, current driving acceleration, and parameter values of future time, future location, future driving speed, and future driving acceleration. The predicted driving trajectory above can be obtained to obtain a more accurate corrected predicted driving trajectory.

In one embodiment, the correction module 708 is further configured to use, when the number of drivable areas in the set of drivable areas is one, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected as the actual driving trajectory of the vehicle to be corrected driving track.

In the driving trajectory determination device in this embodiment, when the number of drivable areas in the drivable area set is one, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected is directly used as the actual driving trajectory of the vehicle to be corrected. If the driving trajectory is determined, the target driving area does not need to be determined, and the driving trajectory correction of the vehicle to be corrected is not required, thereby improving the driving trajectory determination efficiency.

In one embodiment, the first obtaining module 702 is configured to use the target autonomous driving vehicle as a reference to determine a preset distance range corresponding to the target autonomous driving vehicle; add vehicles within the preset distance range to the set of vehicles to be corrected . The correction module 708 is further configured to correct the predicted driving trajectory of the target automatic driving vehicle based on the corrected predicted driving trajectory of each vehicle to be corrected in the set of vehicles to be corrected.

The driving trajectory determination device in this embodiment uses the automatic driving vehicle as a reference to determine a set of vehicles to be corrected located within a preset distance range. Since the predicted driving trajectory of the vehicle to be corrected has been corrected, the predicted driving trajectory of the target autonomous driving vehicle is It should also be corrected to improve the accuracy of the predicted trajectory of the target autonomous vehicle.

For the specific limitation of the driving trajectory determination device, reference may be made to the above limitation on the driving trajectory determination method, which will not be repeated here. All or part of the modules in the above-mentioned driving trajectory determination device can be implemented by software, hardware and combinations thereof. The above modules can be embedded in hardware or independent of the processor in the computer device, or can be stored in the memory of the computer device in software, so that the processor can call and execute the operations corresponding to the above modules.

In one embodiment, a computer device is provided, and the computer device may be a terminal, and its internal structure diagram may be as shown in FIG. 8 . The computer equipment includes a processor, memory, a communication interface, a display screen, and an input device connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The nonvolatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with an external terminal, and the wireless communication can be realized by WIFI, operator network, NFC (Near Field Communication) or other technologies. The computer program, when executed by the processor, implements a driving trajectory determination method. The display screen of the computer equipment may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment may be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the shell of the computer equipment , or an external keyboard, trackpad, or mouse.

Those skilled in the art can understand that the structure shown in FIG. 8 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is also provided, including a memory and a processor, where a computer program is stored in the memory, and the processor implements the steps in the foregoing method embodiments when the processor executes the computer program.

In one embodiment, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, implements the steps in the foregoing method embodiments.

In one embodiment, there is provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the foregoing method embodiments.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage In the medium, when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other media used in the various embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, or optical memory, and the like. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, the RAM may be of various types, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, all It is considered to be the range described in this specification.

The above-mentioned embodiments only represent several embodiments of the present application, and the descriptions thereof are specific and detailed, but should not be construed as a limitation on the scope of the invention patent. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the scope of protection of the patent of the present application shall be subject to the appended claims.

Claims

A driving trajectory determination method, characterized in that the method comprises:

Obtain vehicle driving information of each vehicle to be corrected in the vehicle set to be corrected;

Determine the abnormal state index of each to-be-corrected vehicle based on the vehicle driving information of each to-be-corrected vehicle;

obtaining the environmental information of each vehicle to be corrected in the set of vehicles to be corrected; and

According to the order of the abnormal state indexes in the abnormal state index sequence, the corrected predicted driving trajectory of each vehicle to be corrected is determined based on the environmental information of each vehicle to be corrected; The degree of anomaly is arranged from weak to strong.
The method according to claim 1, wherein the vehicle driving information includes the lateral distance from the center line of the lane where it is located, the current lateral speed of the vehicle, the current driving speed and the current vehicle orientation;

The determining of the abnormal state index of each vehicle to be corrected based on the vehicle driving information of each vehicle to be corrected includes:

Determine the lateral distance deviation index of each vehicle to be corrected based on the lateral distance from the center line of the lane where each vehicle to be corrected is located;

determining a lateral speed deviation index of each to-be-corrected vehicle based on the current vehicle lateral speed of each to-be-corrected vehicle;

determining a travel speed deviation index of each to-be-corrected vehicle based on the current travel speed of each to-be-corrected vehicle;

Determine the orientation deviation index of each vehicle to be corrected based on the current vehicle orientation of each vehicle to be corrected and the orientation of the centerline of the lane where the vehicle to be corrected is located;

An abnormal state index of each vehicle to be corrected is determined based on the lateral distance deviation index, the lateral speed deviation index, the traveling speed deviation index, and the heading deviation index of each vehicle to be corrected.
The method according to claim 2, characterized in that, the determining the driving speed deviation index of each vehicle to be corrected based on the current driving speed of each vehicle to be corrected comprises:

Obtain the lane speed limit corresponding to the lane where the vehicle to be corrected is located;

determining a speed deviation value of each vehicle to be corrected based on the current driving speed and the lane speed limit;

A travel speed deviation index of each vehicle to be corrected is determined based on the speed deviation value.
The method according to claim 1, wherein the environmental information includes environmental area information and environmental vehicle information;

The determining, based on the environmental information of each vehicle to be corrected, the corrected predicted travel trajectory of each vehicle to be corrected includes:

For each vehicle to be corrected, determining a set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected;

determining an area probability value of each drivable area in the drivable area set based on the environmental vehicle information of the vehicle to be corrected;

Determine the target driving area of the vehicle to be corrected based on the area probability value of each drivable area;

A corrected predicted driving trajectory of the vehicle to be corrected is determined based on the target driving area.
The method according to claim 4, wherein the environmental vehicle information comprises a predicted driving trajectory of the environmental vehicle;

The determining an area probability value of each drivable area in the drivable area set based on the environmental vehicle information of the vehicle to be corrected includes:

Searching within a preset distance range of the vehicle to be corrected to obtain an environmental vehicle set of the to-be-corrected vehicle, where the environmental vehicles in the environmental vehicle set have a potential interaction relationship with the to-be-corrected vehicle;

For each drivable area in the set of drivable areas, based on the predicted driving trajectories of the environmental vehicles in the set of environmental vehicles, determine a target trajectory point located in each drivable area;

Determine the area cost value of each drivable area based on the target trajectory points located in each drivable area;

The area probability value corresponding to each drivable area is determined according to the area cost value of each drivable area.
The method according to claim 5, wherein the determining the area cost value of each drivable area based on the target trajectory points located in each drivable area comprises:

Obtain the point weight corresponding to the target trajectory point;

For the predicted trajectory set corresponding to each environmental vehicle, obtain the trajectory probability value corresponding to each predicted trajectory;

For each environmental vehicle, based on the sum of the point weights of the target trajectory points of each predicted driving trajectory and the trajectory probability value of each predicted driving trajectory of the environmental vehicle, the driving generation of the environmental vehicle for the vehicle in the drivable area is obtained. value;

Based on the sum of the vehicle travel cost values of each environmental vehicle in the environmental vehicle set, the area cost value of each drivable area is determined.
The method according to claim 6, wherein the determining the area cost value of each drivable area based on the sum of the vehicle driving cost values of the various environmental vehicles includes:

For each drivable area, obtain the lane-changing cost value of the vehicle to be corrected relative to the drivable area;

obtaining the value of the lateral distance between the center position of the vehicle to be corrected and the center line of the drivable area;

The area cost value of the drivable area is determined based on the lane change cost value, the lateral distance value, and the sum of the vehicle travel cost values of the respective environmental vehicles.
The method according to claim 5, wherein the obtaining the set of environmental vehicles of the vehicle to be corrected by searching within a preset distance range of the vehicle to be corrected comprises:

Taking the vehicle in the first area as the environmental vehicle of the vehicle to be corrected, the set of environmental vehicles corresponding to the first area is obtained; the first area is within the preset distance range and within the to-be-corrected The area directly to the side of the vehicle and the area in the same area as the area where the vehicle to be corrected is located;

The vehicle with the closest distance to the vehicle to be corrected in each second area is used as an environmental vehicle, and an environmental vehicle set corresponding to the second area is obtained; the second area is within the preset distance range except for the first an area outside an area;

The environmental vehicle set corresponding to the first area and the environmental vehicle set corresponding to the second area are used as the environmental vehicle set of the vehicle to be corrected.
The method according to claim 4, wherein the determining the set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected comprises:

Taking the location of the vehicle to be corrected as a starting point, determining the through area and lane change area corresponding to the vehicle to be corrected, and adding the through area and the lane change area to the set of drivable areas of the vehicle to be corrected.
The method according to claim 4, wherein the determining the corrected predicted driving trajectory of the vehicle to be corrected based on the target driving area comprises:

obtaining the current location of the vehicle to be corrected;

obtaining the future location of the vehicle to be corrected, where the future location is located in the target driving area;

Based on the current location and the future location, a corrected predicted travel trajectory of the vehicle to be corrected is determined.
The method according to claim 10, wherein the acquiring the current location of the vehicle to be corrected comprises:

The center line of the target driving area is used as the coordinate axis of the first direction in the coordinate system, and the axis perpendicular to the first direction is used as the coordinate axis of the second direction, and the vehicle driving coordinate system is established;

Obtaining the current location of the vehicle to be corrected in the vehicle traveling coordinate system;

The acquiring the future location of the vehicle to be corrected includes:

obtaining the future location of the vehicle to be corrected in the vehicle traveling coordinate system;

The determining, based on the current location and the future location, the corrected predicted driving trajectory of the vehicle to be corrected includes:

In the vehicle traveling coordinate system, a corrected predicted traveling trajectory of the vehicle to be corrected is determined based on the current location and the future location.
The method according to claim 11, wherein the preset functional relationship includes a first preset functional relationship and a second preset functional relationship, and the first preset functional relationship includes a time parameter and a first preset functional relationship. A position parameter in one direction; the second functional relationship includes a time parameter and a position parameter in the second direction;

Described based on the current location and the future location to determine the corrected predicted travel trajectory of the vehicle to be corrected, including:

Acquiring the current time of the vehicle to be corrected, the corresponding current driving speed, the current driving acceleration, and the future time and the corresponding future driving speed and future driving acceleration;

Inputting the position parameter value corresponding to the first direction in the current position and the current moment into the first preset function relational formula to obtain the first relational formula;

Inputting the first directional component of the current moment and the current driving speed into the first-order derivative of the first preset functional relationship to obtain a second relationship;

Inputting the current moment and the first direction component of the current driving acceleration into the second-order derivative function of the first preset functional relationship to obtain a third relationship;

Inputting the first direction component in the future time and the future driving speed into the first derivative function of the first preset function relational expression to obtain a fourth relational expression;

inputting the first direction component in the future time and the future driving acceleration into the second-order derivative function of the first preset functional relationship to obtain a fifth relationship;

Based on the first relational expression, the second relational expression, the third relational expression, the fourth relational expression, and the fifth relational expression, a revised predicted travel in the first direction is obtained track;

Inputting the position parameter value corresponding to the second direction in the current position and the current moment into the second preset function relational formula to obtain the sixth relational formula;

Inputting the second directional component of the current moment and the current driving speed into the first-order derivative of the second preset functional relationship to obtain a seventh relationship;

Inputting the second direction component of the current moment and the current driving acceleration into the second-order derivative function of the second preset functional relationship to obtain an eighth relationship;

Inputting the second direction component corresponding to the future time and the future position into the second preset functional relationship to obtain the ninth relationship;

Inputting the second direction component in the future time and the future driving speed into the first derivative function of the second preset function relational expression to obtain a tenth relational expression;

Inputting the second direction component of the future time and the future driving acceleration into the second-order derivative function of the second preset functional relationship to obtain an eleventh relationship;

Based on the sixth relational expression, the seventh relational expression, the eighth relational expression, the ninth relational expression, the tenth relational expression and the eleventh relational expression, the second relational expression is obtained The revised predicted driving trajectory in the direction;

A modified predicted travel trajectory of the vehicle to be corrected is obtained based on the modified predicted travel trajectory in the first direction and the modified predicted travel trajectory in the second direction.
The method according to claim 4, wherein the method further comprises:

When the number of drivable areas in the set of drivable areas is one, the predicted driving trajectory corresponding to the drivable area of the vehicle to be corrected is taken as the actual driving trajectory of the vehicle to be corrected.
The method according to any one of claims 1 to 13, wherein the acquisition method of the set of vehicles to be corrected includes:

Using the target autonomous driving vehicle as a reference, determine a preset distance range corresponding to the target autonomous driving vehicle;

adding vehicles within the preset distance range to the set of vehicles to be corrected;

The method also includes:

The predicted travel trajectory of the target autonomous driving vehicle is corrected based on the corrected predicted travel trajectory of each to-be-corrected vehicle in the to-be-corrected vehicle set.
A driving trajectory determination device, characterized in that the device comprises:

a first obtaining module, configured to obtain vehicle driving information of each vehicle to be corrected in the set of vehicles to be corrected;

an abnormal state index determination module, configured to determine the abnormal state index of each vehicle to be corrected based on the vehicle driving information of each vehicle to be corrected;

a second obtaining module, configured to obtain the environmental information of each vehicle to be corrected in the set of vehicles to be corrected;

a correction module, configured to determine the corrected predicted driving trajectory of each vehicle to be corrected based on the environmental information of each vehicle to be corrected according to the order of the abnormal state index in the abnormal state index sequence; The degree of anomaly represented by the state index is arranged from weak to strong.
The device according to claim 15, wherein the vehicle driving information comprises the lateral distance from the center line of the lane where it is located, the current lateral speed of the vehicle, the current driving speed and the current vehicle orientation;

The abnormal state index determination module is configured to determine the lateral distance deviation index of each to-be-corrected vehicle based on the lateral distance of each to-be-corrected vehicle from the lane centerline;

The abnormal state index determination module is configured to determine a lateral speed deviation index of each vehicle to be corrected based on the current lateral speed of each vehicle to be corrected;

The abnormal state index determination module is configured to determine the travel speed deviation index of each vehicle to be corrected based on the current travel speed of each vehicle to be corrected;

The abnormal state index determination module is configured to determine the orientation deviation index of each vehicle to be corrected based on the current vehicle orientation of each vehicle to be corrected and the orientation of the centerline of the lane where the vehicle to be corrected is located;

The abnormal state index determination module is configured to determine the abnormal state index of each vehicle to be corrected based on the lateral distance deviation index, the lateral speed deviation index, the driving speed deviation index and the heading deviation index of each vehicle to be corrected .
The device according to claim 16, wherein the abnormal state index determination module is configured to obtain the lane speed limit corresponding to the lane where each vehicle to be corrected is located; determine the speed limit based on the current driving speed and the lane speed limit The speed deviation value of each vehicle to be corrected; the running speed deviation index of each vehicle to be corrected is determined based on the speed deviation value.
The apparatus according to claim 15, wherein the environmental information includes environmental area information and environmental vehicle information;

The correction module is configured to, for each vehicle to be corrected, determine a set of drivable areas of the vehicle to be corrected based on the environmental area information of the vehicle to be corrected;

The correction module is configured to determine an area probability value of each drivable area in the drivable area set based on the environmental vehicle information of the vehicle to be corrected;

The correction module is configured to determine the target driving area of the vehicle to be corrected based on the area probability value of each drivable area;

The correction module is configured to determine the corrected predicted driving trajectory of the vehicle to be corrected based on the target driving area.
A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein the processor implements the steps of the method according to any one of claims 1 to 14 when the processor executes the computer program.
A computer-readable storage medium on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 14 are implemented.