CN114620070A - Driving track planning method, device, equipment and storage medium - Google Patents

Abstract

The present application provides a driving trajectory planning method, device, equipment and storage medium, and relates to the technical field of intelligent driving. The driving trajectory planning method includes: determining an expected interaction decision between a target vehicle and an obstacle within a preset time range in the future, and based on the expected interaction decision Interactive decision-making, constructing a longitudinal feasible space map of the target vehicle corresponding to a preset time range in the future, obtaining the reference speed of the target vehicle based on the longitudinal feasible space map and the first preset cost function, and determining the target vehicle at a preset time in the future based on the reference speed target trajectory within the range. The present application can obtain the optimal path more accurately, improve the rationality and reliability of the driving trajectory planning result, improve the driving efficiency, and improve the driving experience.

Description

Driving trajectory planning method, device, device and storage medium

technical field

The present application relates to the technical field of intelligent driving, and in particular, to a driving trajectory planning method, device, device and storage medium.

Background technique

With the development of autonomous driving technology, autonomous vehicles are gradually developed and applied. When the self-driving vehicle is driving, a planned driving trajectory will be provided for the self-driving vehicle, so that the self-driving vehicle can automatically drive according to the planned driving trajectory.

At present, the path optimization method based on sampling is often used to plan the driving trajectory of the autonomous vehicle. Determine the optimal path among multiple alternative paths. However, the driving trajectory obtained by the above method may not be optimal.

SUMMARY OF THE INVENTION

The present application provides a driving trajectory planning method, device, device and storage medium to solve the problem that the driving trajectory obtained by the sampling-based path optimization method may not be optimal.

In a first aspect, the present application provides a driving trajectory planning method, including:

Determine the expected interaction decision between the target vehicle and the obstacle within a preset time range in the future. The expected interaction decision is used to characterize the interaction behavior between the target vehicle and the obstacle and the interaction time window corresponding to the interaction behavior. The interaction behavior includes the target vehicle actively passing the obstacle. , the target vehicle actively yields to obstacles and the target vehicle actively ignores obstacles;

Based on the expected interaction decision, construct a longitudinal feasible space map of the target vehicle corresponding to the future preset time range, and the longitudinal feasible space map is used to represent the longitudinal feasible range and longitudinal speed constraint information corresponding to the target vehicle at discrete moments in the future preset time range;

obtaining a reference speed of the target vehicle based on the longitudinal feasible space map and a first preset cost function, where the first preset cost function is determined based on the driving efficiency and comfort of the target vehicle;

Based on the reference speed, the target trajectory of the target vehicle in the future preset time range is determined.

Optionally, based on the expected interaction decision, constructing a longitudinal feasible space map of the target vehicle corresponding to the future preset time range, including: determining, based on the expected interaction decision, the target longitudinal minimum feasible distance corresponding to the target vehicle at discrete moments within the future preset time range , the maximum feasible longitudinal distance of the target, the minimum value of the longitudinal speed of the target, and the maximum value of the longitudinal speed of the target; construct the target according to the minimum feasible distance in the longitudinal direction of the target, the maximum feasible distance in the longitudinal direction of the target, the minimum value of the longitudinal speed of the target, and the maximum value of the longitudinal speed of the target The longitudinal feasible space map of the vehicle corresponding to the preset time range in the future.

Optionally, based on the expected interaction decision, determine the minimum target longitudinal feasible distance, the target longitudinal maximum feasible distance, the minimum target longitudinal speed, and the maximum target longitudinal speed corresponding to the target vehicle at discrete moments within a preset time range in the future, including : Determine the initial maximum feasible longitudinal distance, the minimum initial feasible longitudinal distance, the minimum initial longitudinal velocity, and the maximum initial longitudinal velocity of the target vehicle; for each obstacle corresponding to discrete moments within a preset time range in the future, perform the following operations , until each obstacle is traversed: Based on the expected interaction decision, if it is determined that the target vehicle needs to actively give way to the target obstacle, then according to the space occupied by the target obstacle in the vertical and horizontal directions, the preset safe distance to give way and the initial vertical feasible maximum Distance, determine the target longitudinal maximum feasible distance of the target vehicle, and according to the speed of the target obstacle, determine the maximum longitudinal speed of the target, determine the maximum longitudinal feasible distance of the target as the new initial maximum feasible longitudinal distance, and determine the maximum longitudinal speed of the target is the maximum value of the new initial longitudinal speed; or, based on the expected interaction decision, if it is determined that the target vehicle needs to actively pass the target obstacle, then according to the space occupied by the target obstacle longitudinally and laterally, the preset overtaking safety distance and the initial longitudinal feasibility Minimum distance, determine the target longitudinal feasible minimum distance of the target vehicle, and determine the minimum target longitudinal speed according to the speed of the target obstacle, determine the target longitudinal feasible minimum distance as the new initial longitudinal feasible minimum distance, and determine the minimum target longitudinal speed. The value is the minimum value of the new initial longitudinal velocity.

Optionally, determining the initial longitudinal maximum feasible distance of the target vehicle includes: determining the maximum feasible distance of the target vehicle within a preset time range in the future based on the speed limit of the driving scene of the target vehicle; Determine the maximum speed corresponding to the maximum curvature corresponding to the maximum curvature of the traveled path and the corresponding relationship between the preset curvature and the speed constraint; obtain the maximum feasible distance of the target according to the maximum speed and the maximum feasible distance; determine the maximum feasible distance of the target as the initial longitudinal feasible maximum distance .

Optionally, after constructing a longitudinal feasible space map of the target vehicle corresponding to a preset time range in the future, the driving trajectory planning method further includes: aiming at the minimum longitudinal feasible distance of the target and the maximum feasible longitudinal distance of the target corresponding to discrete moments in the predetermined time range in the future. , according to the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment, obtain the updated target longitudinal feasible minimum distance corresponding to the current discrete moment, and obtain the updated target longitudinal feasible minimum distance corresponding to the current discrete moment, and according to the target maximum feasible distance and the current discrete moment The corresponding maximum vertical feasible distance of the target, and the updated maximum vertical feasible distance of the target corresponding to the current discrete moment is obtained; according to the updated minimum vertical feasible distance of the target and the updated maximum vertical feasible distance of the target, the updated vertical feasible space map is obtained. .

Optionally, according to the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment, the updated target longitudinal feasible minimum distance corresponding to the current discrete moment is obtained, and the target maximum feasible distance and Obtaining the updated maximum longitudinal feasible distance of the target corresponding to the current discrete moment, and obtaining the updated maximum longitudinal feasible distance of the target corresponding to the current discrete moment, including: for discrete moments within a preset time range in the future, if it is determined that the updated maximum longitudinal feasible distance of the target is less than the updated target After the target longitudinal feasible minimum distance, the interaction behavior between the corresponding obstacle and the target vehicle is updated, and the longitudinal feasible space map of the target vehicle corresponding to the future preset time range is reconstructed.

Optionally, based on the longitudinal feasible space map and the first preset cost function, obtain the reference speed of the target vehicle, including: based on the longitudinal feasible space map, according to the maximum acceleration capability, maximum deceleration capability of the target vehicle, and the path speed within the longitudinal feasible range. The curvature corresponding to the reference line and the corresponding relationship between the preset curvature and the speed constraint, obtain the updated longitudinal feasible range and longitudinal speed constraint information; obtain the updated longitudinal feasible space according to the updated longitudinal feasible range and longitudinal speed constraint information Figure; based on the updated longitudinal feasible space map and the first preset cost function, obtain the first target position of the target vehicle corresponding to the discrete time; perform fitting processing on the first target position to obtain the reference speed of the target vehicle.

Optionally, based on the reference speed, determining the target trajectory of the target vehicle in the future preset time range includes: obtaining the target candidate path corresponding to the future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed , the preset candidate path is obtained by vertically and horizontally sampling the future travel path of the target vehicle; according to the target candidate path and the second preset cost function, the target trajectory of the target vehicle within the preset time range in the future is determined, and the second preset path is determined. The cost function is assumed to be determined based on the driving safety, comfort and stability of the target vehicle.

Optionally, obtaining a target candidate path corresponding to a preset time range in the future according to at least one preset candidate path corresponding to the target vehicle and the reference speed, including: for each preset candidate path, according to the reference speed, obtaining a discrete time corresponding to the speed. The longitudinal advance distance corresponding to the target vehicle; according to the longitudinal advance distance and the preset candidate path, the second target position of the target vehicle corresponding to the discrete time is obtained; the coordinate transformation of the second target position is performed to obtain the target candidate path.

In a second aspect, the present application provides a driving trajectory planning device, comprising:

The determination module is used to determine the expected interaction decision between the target vehicle and the obstacle within a preset time range in the future. The expected interaction decision is used to characterize the interaction behavior between the target vehicle and the obstacle and the interaction time window corresponding to the interaction behavior. The interaction behavior includes the target The vehicle actively overtakes the obstacle, the target vehicle actively yields to the obstacle, and the target vehicle actively ignores the obstacle;

The building module is used to construct a longitudinal feasible space map of the target vehicle corresponding to a preset time range in the future based on the expected interaction decision. speed constraint information;

an obtaining module, configured to obtain the reference speed of the target vehicle based on the longitudinal feasible space map and a first preset cost function, where the first preset cost function is determined based on the driving efficiency and comfort of the target vehicle;

The processing module is used for determining the target trajectory of the target vehicle in the future preset time range based on the reference speed.

Optionally, the building module is specifically used to: determine, based on the expected interaction decision, the minimum target longitudinal feasible distance, the target longitudinal maximum feasible distance, the target longitudinal speed minimum, and the target longitudinal distance corresponding to the target vehicle at discrete moments within a preset time range in the future. The maximum value of the speed; according to the minimum feasible longitudinal distance of the target, the maximum feasible longitudinal distance of the target, the minimum value of the target longitudinal speed and the maximum value of the target longitudinal speed, the longitudinal feasible space map of the target vehicle corresponding to the preset time range in the future is constructed.

Optionally, the building module is used to determine, based on the expected interaction decision, the minimum target longitudinal feasible distance, the target longitudinal maximum feasible distance, the minimum target longitudinal velocity, and the target longitudinal velocity corresponding to the target vehicle at discrete moments within a preset time range in the future. When the maximum value of the initial longitudinal speed is determined, it is specifically used to: determine the initial maximum feasible longitudinal distance of the target vehicle, the initial minimum feasible longitudinal distance, the minimum initial longitudinal speed, and the maximum initial longitudinal speed; For each obstacle, perform the following operations until each obstacle is traversed: Based on the expected interaction decision, if it is determined that the target vehicle needs to actively yield to the target obstacle, the space occupied by the target obstacle longitudinally and laterally, and the preset yield The safety distance and the initial maximum feasible longitudinal distance, determine the target maximum longitudinal feasible distance of the target vehicle, and according to the speed of the target obstacle, determine the maximum longitudinal speed of the target, and determine the maximum feasible longitudinal distance of the target as the new initial maximum feasible longitudinal distance , determine the maximum value of the target longitudinal speed as the new maximum value of the initial longitudinal speed; or, based on the expected interaction decision, if it is determined that the target vehicle needs to actively surpass the target obstacle, then the space occupied by the target obstacle longitudinally and laterally, Set the overtaking safety distance and the initial minimum feasible longitudinal distance, determine the target minimum longitudinal feasible distance of the target vehicle, and determine the minimum value of the target longitudinal speed according to the speed of the target obstacle, and determine the target longitudinal feasible minimum distance as the new initial longitudinal feasible minimum distance. distance, and determine the minimum value of the target longitudinal velocity as the minimum value of the new initial longitudinal velocity.

Optionally, when the building module is used to determine the initial longitudinal feasible maximum distance of the target vehicle, it is specifically used to: determine the maximum feasible distance of the target vehicle within a preset time range in the future based on the speed limit of the driving scene of the target vehicle; The maximum curvature of the path traveled by the vehicle within the preset time range in the future and the corresponding relationship between the preset curvature and the speed constraint, determine the maximum speed corresponding to the maximum curvature; obtain the maximum feasible distance of the target according to the maximum speed and the maximum feasible distance; determine the maximum possible distance of the target The feasible distance is the initial longitudinal feasible maximum distance.

Optionally, the driving trajectory planning device further includes an update module for, after the building module constructs the longitudinal feasible space map of the target vehicle corresponding to the future preset time range, for the target longitudinal feasible minimum corresponding to the discrete moments in the future preset time range. distance and the target longitudinal feasible maximum distance, according to the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment, obtain the updated target longitudinal feasible minimum distance corresponding to the current discrete moment, and according to the target The maximum feasible distance and the maximum vertical feasible distance of the target corresponding to the current discrete time are obtained, and the updated maximum vertical feasible distance of the target corresponding to the current discrete time is obtained; according to the updated minimum vertical feasible distance and the updated maximum vertical feasible distance of the target, obtain The updated vertical feasible space map.

Optionally, the update module is used to obtain the updated target longitudinal feasible minimum distance corresponding to the current discrete time according to the target longitudinal feasible minimum distance corresponding to the previous discrete time and the target longitudinal feasible minimum distance corresponding to the current discrete time, and according to The maximum feasible distance of the target and the maximum vertical feasible distance of the target corresponding to the current discrete time, and when the updated maximum vertical feasible distance of the target corresponding to the current discrete time is obtained, it is specifically used: for the discrete time within the preset time range in the future, if it is determined that after the update If the target longitudinal feasible maximum distance is less than the updated target longitudinal feasible minimum distance, the interaction behavior between the corresponding obstacle and the target vehicle is updated, and the longitudinal feasible space map of the target vehicle corresponding to the future preset time range is reconstructed.

Optionally, the acquisition module is specifically used for: based on the longitudinal feasible space map, according to the maximum acceleration capability, the maximum deceleration capability of the target vehicle, the curvature corresponding to the reference line of the path within the longitudinal feasible range, and the corresponding relationship between the preset curvature and the speed constraint. , obtain the updated longitudinal feasible range and longitudinal speed constraint information; obtain the updated longitudinal feasible space map according to the updated longitudinal feasible range and longitudinal speed constraint information; based on the updated longitudinal feasible space map and the first preset cost function to obtain the first target position of the target vehicle corresponding to the discrete moment; perform fitting processing on the first target position to obtain the reference speed of the target vehicle.

Optionally, the processing module is specifically configured to: obtain a target candidate path corresponding to a preset time range in the future according to at least one preset candidate path corresponding to the target vehicle and the reference speed, and the preset candidate path is to perform a separate operation on the future travel path of the target vehicle. Obtained by vertical and horizontal sampling; according to the target candidate path and the second preset cost function, the target trajectory of the target vehicle in the future preset time range is determined, and the second preset cost function is based on the driving safety and comfort of the target vehicle. and stability is determined.

Optionally, when the processing module is used to obtain the target candidate path corresponding to the future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed, the processing module is specifically used for: for each preset candidate path, according to With reference to the speed, the longitudinal advance distance corresponding to the target vehicle corresponding to the discrete time is obtained; according to the longitudinal advance distance and the preset candidate path, the second target position of the target vehicle corresponding to the discrete time is obtained; the coordinate transformation of the second target position is performed to obtain the target candidate path.

In a third aspect, the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

memory stores instructions for execution by the computer;

The processor executes the computer-executed instructions stored in the memory to implement the driving trajectory planning method according to the first aspect of the present application.

In a fourth aspect, the present application provides a computer-readable storage medium, where computer program instructions are stored in the computer-readable storage medium, and when the computer program instructions are executed by a processor, the driving trajectory planning method described in the first aspect of the present application is implemented. .

In a fifth aspect, the present application provides a computer program product, including a computer program, when the computer program is executed by a processor, the driving trajectory planning method according to the first aspect of the present application is implemented.

The driving trajectory planning method, device, device and storage medium provided by the present application, by determining the expected interaction decision between the target vehicle and the obstacle in the future preset time range, and based on the expected interaction decision, construct the target vehicle corresponding to the future preset time range. The longitudinal feasible space map, based on the longitudinal feasible space map and the first preset cost function, obtains the reference speed of the target vehicle, and determines the target trajectory of the target vehicle in the future preset time range based on the reference speed. Because this application fully considers the information of static obstacles and dynamic obstacles in the driving scene, determines the desired interaction decision, and then constructs the longitudinal feasible space map of the target vehicle, and obtains the reference speed closest to the final execution of the target vehicle based on the longitudinal feasible space map, The reference speed is used to evaluate the optimal path, so the optimal path can be obtained more accurately, the rationality and reliability of the driving trajectory planning result are improved, the driving efficiency is improved, and the driving experience is improved.

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 will briefly introduce the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present application, and for those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application;

2 is a flowchart of a driving trajectory planning method provided by an embodiment of the present application;

3 is a flowchart of a driving trajectory planning method provided by another embodiment of the present application;

FIG. 4 is a schematic structural diagram of a driving trajectory planning device provided by an embodiment of the present application;

FIG. 5 is a schematic structural diagram of an electronic device provided by the present application.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of the financial data or user data involved are all in compliance with relevant laws and regulations, and do not violate public order and good customs.

At present, in the driving trajectory planning scheme of automatic driving, the path and speed decoupling planning method is mostly used, and a set of safe and flexible path planning strategy needs to fully consider the static obstacles and dynamic obstacles in the driving scene. , that is, at a certain moment in the future, whether the self-driving vehicle (that is, the current self-driving vehicle) and the obstacle will appear in the same position at the same time. Therefore, when evaluating alternative paths, it is necessary to provide corresponding speed and time information for discrete waypoints on the path to analyze whether there is a potential collision risk with obstacles. For the path optimization method based on sampling, unreasonable reference speed will lead to evaluation deviation, and then the optimal driving path cannot be selected, which affects driving safety and driving efficiency.

Based on the above problems, the present application provides a driving trajectory planning method, device, equipment and storage medium. By fully considering the information of static obstacles and dynamic obstacles in the driving scene, based on the behavior and trajectory prediction results of obstacles, combined with The execution ability of the target vehicle, under the premise of reasonable interaction with obstacles, optimize the reference speed, so as to obtain the reference speed closest to the final execution of the target vehicle, and use the reference speed for optimal path evaluation, which can be more accurate It can obtain the optimal path and improve the rationality and reliability of the driving trajectory planning results.

Hereinafter, the application scenarios of the solutions provided by the present application are firstly described with examples.

FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application. As shown in FIG. 1 , in this application scenario, the autonomous vehicle 101 drives on the road 102 according to the planned driving trajectory. The specific implementation process of how the automatic driving vehicle 101 obtains the planned driving trajectory may refer to the solutions of the following embodiments.

It should be noted that FIG. 1 is only a schematic diagram of an application scenario provided by the embodiment of the present application, and the embodiment of the present application does not limit the devices included in FIG. 1 , nor does it limit the positional relationship between the devices in FIG. 1 .

Next, the driving trajectory planning method is introduced through specific embodiments.

FIG. 2 is a flowchart of a driving trajectory planning method provided by an embodiment of the present application. The methods in the embodiments of the present application may be applied to an electronic device, and the electronic device may be a server or a server cluster, or the like. As shown in FIG. 2, the method of the embodiment of the present application includes:

S201. Determine the expected interaction decision between the target vehicle and the obstacle within a preset time range in the future.

Among them, the expected interaction decision is used to characterize the interaction behavior between the target vehicle and the obstacle and the interaction time window corresponding to the interaction behavior. The interaction behavior includes the target vehicle actively overtaking the obstacle, the target vehicle actively yielding to the obstacle, and the target vehicle actively ignoring the obstacle.

In the embodiment of the present application, for example, the preset time range in the future is, for example, 10s in the future. Based on the obstacle behavior and trajectory prediction results given by the preset prediction module, taking into account information such as public road driving rules, road rights and speed limits, preliminary interaction can be performed for each obstacle that has potential interaction with the target vehicle Decision-making, that is, to determine the expected interaction decision of the target vehicle with the obstacle within a preset time range in the future. Specifically, the interaction behavior between the target vehicle and the obstacle can also be called interaction attribute, and the interaction attribute is defined in three types: the target vehicle needs to actively pass the obstacle (denoted as pass_type), the target vehicle needs to actively yield to the obstacle (denoted as pass_type) yield_type) and obstacles (denoted ignore_type) that are actively ignored because they do not pose a potential risk to the target vehicle. The above three interaction attributes can be understood as decision-making at the behavioral level.

For the obstacle that interacts with the target vehicle, in addition to the decision at the behavior level, it is also necessary to give the interaction time window corresponding to the decision at the behavior level, that is, the corresponding overtaking or yield time window. The value range of the interaction time window can be Expressed as [t_min, t_max], where t_min represents the minimum start time of the interaction time window, and t_max represents the maximum end time of the interaction time window. For example, if it is expected to overtake an obstacle whose identity document (id) is represented by obs_0 within 3s to 5s, the interaction attribute of the obstacle is marked as pass_type, and the corresponding interaction time window is [3,5] ; For example, it is expected to yield to an obstacle whose id is represented by obs_1 within 8s to 10s, then mark the interaction attribute of the obstacle as yield_type, and the corresponding interaction time window is [8, 10]; if an id is The obstacle of obs_2 will not drive into the lane where the target vehicle is located within 10s, or there is no potential interaction, then mark the interaction attribute of the obstacle as ignore_type, and the corresponding interaction time window is [0,10]. The interaction analysis of all obstacles in the driving scene of the target vehicle is performed in turn, and the expected interaction decision between the target vehicle and each obstacle can be obtained.

S202. Based on the expected interaction decision, construct a longitudinal feasible space map of the target vehicle corresponding to a preset time range in the future.

Among them, the longitudinal feasible space map is used to represent the longitudinal feasible range and longitudinal speed constraint information corresponding to the target vehicle at discrete moments in the future preset time range.

In this step, after the expected interaction decision is obtained, a longitudinal feasible space map of the target vehicle corresponding to the future preset time range can be constructed based on the expected interaction decision, so as to determine the longitudinal feasible space map of the target vehicle corresponding to discrete moments in the future preset time range. Range and longitudinal velocity constraint information. Exemplarily, the abscissa of the vertical feasible space map is, for example, discrete moments within a preset time range in the future, and the ordinate of the vertical feasible space map is, for example, the vertical feasible range corresponding to each discrete moment, and there is a corresponding vertical feasible range. longitudinal velocity constraint information. For how to construct a longitudinal feasible space map of the target vehicle corresponding to a predetermined time range in the future based on the expected interaction decision, reference may be made to the subsequent embodiments, which will not be repeated here.

S203. Obtain a reference speed of the target vehicle based on the longitudinal feasible space map and the first preset cost function.

The first preset cost function is determined based on the driving efficiency and comfort of the target vehicle.

Exemplarily, too low driving speed will bring about the problem of low driving efficiency, while too fast speed will easily affect the driving experience, and even cause vehicle instability in severe cases. Therefore, a first preset cost function (expressed as cost_function), it is necessary to weigh the driving efficiency and comfort of the target vehicle, that is, for the consideration of driving efficiency, it is hoped that the driving distance tends to be as far as possible; The jerk (ie, the increment of acceleration) tends to be as small as possible. For the specific definition of the first preset cost function, reference may be made to subsequent embodiments, which will not be repeated here. In this step, after the longitudinal feasible space map is obtained, the reference speed of the target vehicle may be obtained based on the longitudinal feasible space map and the first preset cost function. For how to obtain the reference speed of the target vehicle based on the longitudinal feasible space map and the first preset cost function, reference may be made to subsequent embodiments, which will not be repeated here.

S204. Based on the reference speed, determine the target trajectory of the target vehicle within a preset time range in the future.

In this step, after the reference speed is obtained, the target trajectory of the target vehicle in the future preset time range may be determined based on the reference speed. For how to determine the target trajectory of the target vehicle in the future preset time range based on the reference speed, reference may be made to subsequent embodiments, which will not be repeated here.

After the target trajectory of the target vehicle in the future preset time range is determined, the optimal trajectory of the target vehicle in the future preset time range is determined, and the target vehicle can be controlled to travel according to the planned optimal trajectory.

In the driving trajectory planning method provided by the embodiment of the present application, by determining the expected interaction decision between the target vehicle and the obstacle in the future preset time range, and based on the expected interaction decision, a longitudinal feasible space map of the target vehicle corresponding to the future preset time range is constructed, Based on the longitudinal feasible space map and the first preset cost function, the reference speed of the target vehicle is obtained, and based on the reference speed, the target trajectory of the target vehicle in the future preset time range is determined. Since the embodiment of the present application fully considers the information of static obstacles and dynamic obstacles in the driving scene, the desired interaction decision is determined, and then the longitudinal feasible space map of the target vehicle is constructed, and the reference that is closest to the final execution of the target vehicle is obtained based on the longitudinal feasible space map. The reference speed is used to evaluate the optimal path. Therefore, the optimal path can be obtained more accurately, the rationality and reliability of the driving trajectory planning result can be improved, the driving efficiency can be improved, and the driving experience can be improved.

FIG. 3 is a flowchart of a driving trajectory planning method provided by another embodiment of the present application. On the basis of the above embodiments, the embodiments of the present application further describe how to plan the driving trajectory. As shown in FIG. 3 , the method of this embodiment of the present application may include:

S301. Determine the expected interaction decision between the target vehicle and the obstacle within a preset time range in the future.

Among them, the expected interaction decision is used to characterize the interaction behavior between the target vehicle and the obstacle and the interaction time window corresponding to the interaction behavior. The interaction behavior includes the target vehicle actively overtaking the obstacle, the target vehicle actively yielding to the obstacle, and the target vehicle actively ignoring the obstacle.

For a specific description of this step, reference may be made to the relevant description of S201 in the embodiment shown in FIG. 2 , and details are not repeated here.

In this embodiment of the present application, step S202 in FIG. 2 may further include the following two steps of S302 and S303:

S302. Determine, based on the expected interaction decision, the minimum target longitudinal feasible distance, the maximum target longitudinal feasible distance, the minimum target longitudinal velocity, and the maximum target longitudinal velocity corresponding to the target vehicle at discrete moments in the future preset time range.

In this step, the expected interaction decision can provide a reference for the construction of the longitudinal feasible space of the target vehicle at each discrete moment in the future preset time range. After obtaining the expected interaction decision between the target vehicle and the obstacle in the future preset time range, it is possible to determine the target longitudinal feasible minimum distance and the target longitudinal feasible distance corresponding to the target vehicle at discrete moments in the future preset time range based on the expected interaction decision. Maximum distance, minimum target longitudinal velocity, and maximum target longitudinal velocity.

Further, optionally, based on the expected interaction decision-making, determine the target longitudinal minimum feasible distance, the target longitudinal maximum feasible distance, the minimum target longitudinal speed, and the maximum target longitudinal speed corresponding to the target vehicle at discrete moments in the future preset time range. value, which may include: determining the initial maximum feasible longitudinal distance of the target vehicle, the initial minimum feasible longitudinal distance, the minimum initial longitudinal speed, and the maximum initial longitudinal speed; for each obstacle corresponding to discrete moments in the future preset time range , and perform the following operations until each obstacle is traversed: based on the expected interaction decision, if it is determined that the target vehicle needs to actively yield to the target obstacle, the space occupied by the target obstacle in the longitudinal and lateral directions, the preset yield safety distance, and Determine the maximum longitudinal feasible distance of the target vehicle, and determine the maximum longitudinal speed of the target according to the speed of the target obstacle, determine the maximum feasible longitudinal distance of the target as the new initial maximum feasible longitudinal distance, and determine the maximum longitudinal feasible distance of the target. The maximum value of the speed is the maximum value of the new initial longitudinal speed; or, based on the expected interaction decision, if it is determined that the target vehicle needs to actively pass the target obstacle, the overtaking safety distance is preset according to the space occupied by the target obstacle longitudinally and laterally. and the initial longitudinal feasible minimum distance, determine the target longitudinal feasible minimum distance of the target vehicle, and determine the minimum target longitudinal speed according to the speed of the target obstacle, determine the target longitudinal feasible minimum distance as the new initial longitudinal feasible minimum distance, and determine the target The minimum longitudinal velocity is the new initial minimum longitudinal velocity.

Wherein, optionally, determining the initial maximum feasible longitudinal distance of the target vehicle may include: determining the maximum feasible distance of the target vehicle within a preset time range in the future based on the speed limit of the driving scene of the target vehicle; Determine the maximum speed corresponding to the maximum curvature corresponding to the maximum curvature of the traveled path within the time range and the corresponding relationship between the preset curvature and the speed constraint; obtain the maximum feasible distance of the target according to the maximum speed and the maximum feasible distance; determine the maximum feasible distance of the target as the initial longitudinal direction The maximum distance possible.

Exemplarily, the preset time range in the future is, for example, 10 s in the future, and the corresponding relationship between the preset curvature and the velocity constraint is, for example, a curvature-velocity constraint table calibrated offline in advance. Exemplarily, the maximum drivable distance of the target vehicle (represented as max_drivable_length) in a non-interference state is obtained, that is, the maximum feasible distance of the target vehicle in the next 10s under physical constraints, and the calculation process fully considers the speed limit and Lane shape limits driving speed. First, considering the speed limit of the driving scene of the target vehicle, initialize max_drivable_length according to the speed limit of the driving scene of the target vehicle, and obtain the initial value of max_drivable_length; secondly, considering the driving stability, the maximum speed of the target vehicle is affected by the shape of the lane, for example, suppose the target When the vehicle enters a right-angle turn section, it needs to decelerate according to the curvature of the path. Based on the offline calibration curvature-speed constraint table and the maximum curvature of a future path, query the curvature-speed constraint table to obtain the corresponding maximum speed. The maximum speed and the initial value of max_drivable_length update max_drivable_length, that is, the maximum feasible distance of the target is obtained.

The maximum feasible distance of the target is determined as the initial maximum feasible longitudinal distance, the minimum initial feasible longitudinal distance is, for example, 0, the minimum value of the initial longitudinal speed is, for example, 0, and the maximum value of the initial longitudinal speed is, for example, the speed limit of the driving scene of the target vehicle. After obtaining the initial maximum feasible longitudinal distance of the target vehicle, the minimum feasible initial longitudinal distance, the minimum initial longitudinal speed, and the maximum initial longitudinal speed, the future preset time range is discretized at equal time intervals to obtain the corresponding At discrete moments, it is necessary to comprehensively consider the reasonable interaction between the target vehicle and each obstacle at each discrete moment to obtain the boundary value of the target longitudinal feasible distance of the target vehicle (ie, the longitudinal feasible range). The boundary value includes the target longitudinal feasible minimum distance (denoted as s_ego_min) and the target longitudinal feasible maximum distance (denoted as s_ego_max), wherein the boundary value of the target longitudinal feasible distance corresponding to each discrete moment needs to satisfy the basic constraints, that is, s_ego_min>=0 , and s_ego_max<=max_drivable_length.

Specifically, for each obstacle corresponding to each discrete time in the future preset time range, taking the discrete time t0 as an example, for each obstacle (represented as obs_i) in the target vehicle driving scene, first, query the corresponding The space occupied by the obstacle longitudinally and laterally in the obstacle trajectory prediction result (represented as sdboundary) information, sdboundary is a rectangular box, which is used to represent the lane position space occupied by the obstacle at a discrete moment, and the longitudinal occupied by the target obstacle. The range of the space is: [obs_i_s_min, obs_i_s_max], where obs_i_s_min represents the minimum value of the space occupied by the target obstacle vertically, obs_i_s_max represents the maximum value of the space occupied by the target obstacle vertically; the space occupied by the target obstacle horizontally The range is: [obs_i_d_min, obs_i_d_max], where obs_i_d_min represents the minimum value of the space occupied by the target obstacle horizontally, obs_i_d_max represents the maximum value of the space occupied by the target obstacle horizontally, and the target vehicle cannot pass through the information according to the sdboundary information. . Next, consider the desired interaction decision of the target vehicle with the obstacle:

If it is necessary to give way to an obstacle, in order to ensure that the target vehicle has enough response time and space in an emergency, the maximum longitudinal feasible distance of the target vehicle should be considered on the basis of the obs_i_s_min occupied by the obstacle to reserve a safe way to go (expressed as lon_buffer), the longitudinal feasible maximum distance of the target vehicle is related to the speed of the obstacle, so the target longitudinal feasible maximum distance of the target vehicle corresponding to t0 (represented as s_ego_max_t0) is updated to the smaller value of s_ego_max_t0 and (obs_i_s_min-lon_buffer); if is an obstacle closer to the target vehicle, then the maximum value of the target longitudinal velocity of the target vehicle (represented as v_ego_max_t0) needs to be updated to the velocity of the obstacle;

If an obstacle needs to be overtaken, the longitudinal safety distance reserved for overtaking should be considered on the basis of the obs_i_s_max occupied by the obstacle. The longitudinal safety distance is related to the speed of the obstacle, so the target vehicle corresponding to the target vehicle when interacting with the obstacle obs_i Longitudinal feasible minimum distance (represented as s_ego_min_t0) is updated to the larger of s_ego_min_t0 and (obs_i_s_max+lon_buffe); if it is an obstacle closer to the target vehicle, the minimum value of the target longitudinal speed of the target vehicle (represented as v_ego_min_t0) is updated to the velocity of the obstacle.

Traverse all the obstacles that need to give way and need to be overtaken, the boundary value of the longitudinal feasible range of the target vehicle corresponding to t0 can be updated, and the boundary value includes the target longitudinal feasible minimum distance and the target longitudinal feasible maximum distance, that is, the longitudinal feasible distance of the target vehicle The range is [s_ego_min_t0, s_ego_max_t0], along with the obstacle id corresponding to the boundary value and the longitudinal speed constraint information of the target vehicle. The longitudinal speed constraint information includes the minimum value of the target longitudinal speed and the maximum value of the target longitudinal speed, namely the target vehicle’s The longitudinal speed constraint range is [v_ego_min_t0,v_ego_max_t0].

Referring to the execution steps at discrete time t0, the longitudinal feasible range of the target vehicle from the 0s to the 10s in the future (corresponding to the minimum feasible longitudinal distance of the target and the maximum feasible longitudinal distance of the target) and the longitudinal speed constraint information (corresponding to the minimum longitudinal speed of the target) can be obtained. value and the maximum value of the target longitudinal velocity).

S303 , constructing a longitudinal feasible space map of the target vehicle corresponding to a preset time range in the future according to the minimum target longitudinal feasible distance, the target longitudinal maximum feasible distance, the minimum target longitudinal velocity, and the maximum target longitudinal velocity.

In this step, for example, the preset time range in the future is, for example, 10 seconds in the future. After obtaining the minimum feasible longitudinal distance of the target, the maximum feasible longitudinal distance of the target, the minimum longitudinal speed of the target, and the maximum longitudinal speed of the target, the The minimum longitudinal feasible distance of the target, the maximum feasible longitudinal distance of the target, the minimum value of the target longitudinal speed, and the maximum value of the target longitudinal speed are constructed, and the longitudinal feasible space map corresponding to the target vehicle in the next 10s is constructed to describe each discrete time of the target vehicle in the next 10s. The longitudinal feasible minimum distance and the longitudinal feasible maximum distance corresponding to the time, that is, the longitudinal feasible range corresponding to each discrete moment of the target vehicle in the next 10s, and the corresponding minimum target longitudinal speed and maximum target longitudinal speed within the longitudinal feasible range. , namely the longitudinal velocity constraint information.

S304. For the target longitudinal feasible minimum distance and the target longitudinal maximum feasible distance corresponding to discrete moments in the future preset time range, obtain the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment. The updated target longitudinal feasible minimum distance corresponding to the current discrete moment, and the updated target longitudinal feasible maximum distance corresponding to the current discrete moment is obtained according to the target maximum feasible distance and the target longitudinal feasible maximum distance corresponding to the current discrete moment.

This step can be understood as the rationality check of the boundary value of the longitudinal feasible range corresponding to the discrete moments in the future preset time range. Exemplarily, since it is not a trajectory planning for reversing, it is necessary to ensure that the target longitudinal feasible minimum distance corresponding to the current discrete moment is not less than the target longitudinal feasible minimum distance corresponding to the adjacent previous discrete moment, and the driving process is constrained by the target maximum feasible distance. .

In this step, further, optionally, according to the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment, the updated target longitudinal feasible minimum distance corresponding to the current discrete moment is obtained, and obtaining the updated maximum longitudinal feasible distance of the target corresponding to the current discrete moment according to the maximum feasible distance of the target and the maximum feasible longitudinal distance of the target corresponding to the current discrete moment, which may include: for discrete moments within a preset time range in the future, if it is determined that after the update If the target longitudinal feasible maximum distance is less than the updated target longitudinal feasible minimum distance, the interaction behavior between the corresponding obstacle and the target vehicle is updated, and the longitudinal feasible space map of the target vehicle corresponding to the future preset time range is reconstructed.

In the above process of checking the rationality of the boundary value of the vertical feasible range, if s_ego_max < s_ego_min at a discrete moment, it means that if the expected interactive decision-making needs to be satisfied, that is, exceeding a certain obstacle will cause the obstacle to be encountered next. If the safety is the highest consideration, it is necessary to modify the expected interaction decision of the obstacle of the pass_type to yield_type, and repeat the above steps S302, S303 and S304 until the boundary values of the vertical feasible range corresponding to all discrete moments are all is a reasonable value, and reconstructs the longitudinal feasible space map of the target vehicle corresponding to the preset time range in the future.

S305. Obtain an updated vertical feasible space map according to the updated minimum vertical feasible distance of the target and the updated maximum vertical feasible distance of the target.

In this step, after obtaining the updated minimum vertical feasible distance of the target and the updated maximum vertical feasible distance of the target corresponding to each discrete moment in the future preset time range, the updated minimum vertical feasible distance of the target and the updated maximum vertical distance of the target can be obtained. Afterwards, the maximum feasible vertical distance of the target is obtained, and the updated vertical feasible space map is obtained.

In this embodiment of the present application, step S203 in FIG. 2 may further include the following four steps S306 to S309:

S306. Based on the longitudinal feasible space map, obtain the updated longitudinal feasible range according to the maximum acceleration capability, the maximum deceleration capability of the target vehicle, the curvature corresponding to the reference line of the path within the longitudinal feasible range, and the corresponding relationship between the preset curvature and the speed constraint and longitudinal velocity constraint information.

In this step, the reference line of the path within the longitudinal feasible range is, for example, the center line of the path within the longitudinal feasible range. Exemplarily, first determine the following three constraints: (1) Considering the driving experience and the continuity of the controller control signal of the target vehicle (such as the control signal corresponding to the opening of the accelerator pedal), it is necessary to ensure the initial state of the target vehicle (ie, (2) Consider the execution capability of the target vehicle, such as the maximum acceleration capability and maximum deceleration capability of the target vehicle, Determine the constraint information of the executable speed at the next discrete moment and the constraint information of the executable distance at the next discrete moment; (3) Consider the shape of the lane, for example, excessive vehicle speed during the turning process may easily cause the target vehicle to become unstable or reduce the ride. Therefore, it is necessary to add constraints to the speed of the target vehicle according to the curvature of the lane line; specifically, for example, for discrete time t, any sampling point that satisfies the boundary value constraint of the longitudinal feasible range is expressed as (t_i, s_i), where, s_i represents the longitudinal feasible distance of any sampling point, query the corresponding curvature information on the reference line, and obtain the maximum speed constraint at this position by querying the corresponding relationship between the preset curvature and the speed constraint.

Based on the longitudinal feasible range and longitudinal speed constraint information corresponding to the target vehicle at discrete moments in the future preset time range in the longitudinal feasible space map, and according to the above three constraints, the updated longitudinal feasible range and longitudinal speed constraint information can be obtained.

S307. Obtain an updated longitudinal feasible space map according to the updated longitudinal feasible range and longitudinal speed constraint information.

In this step, after obtaining the updated longitudinal feasible range and longitudinal speed constraint information corresponding to each discrete moment in the future preset time range, the updated longitudinal feasible range and longitudinal speed constraint information can be obtained according to the updated longitudinal feasible range and longitudinal speed constraint information. Vertical Feasible Space Diagram.

S308 , based on the updated longitudinal feasible space map and the first preset cost function, obtain the first target position of the target vehicle corresponding to the discrete time.

Exemplarily, the first preset cost function cost_function is defined as:

cost_function=coefficient_vel*(max_vel_i–vel_i)+coefficient_acc*acc_i+coefficient_jerk*jerk_i

Among them, coefficient_vel, coefficient_acc, and coefficient_jerk represent the velocity weighting factor, acceleration weighting factor, and jerk weighting factor of the first preset cost function, respectively, and are all positive values; vel_i represents the velocity of the i-th discrete time sampling point, max_vel_i The maximum velocity determined according to the maximum curvature of the path at the ith discrete time; acc_i represents the acceleration of the sampling point at the ith discrete time; jerk_i represents the jerk at the sampling point at the ith discrete time.

In this step, after the updated vertical feasible space map is obtained, the first target position of the target vehicle corresponding to each discrete moment can be obtained based on the updated vertical feasible space map and the first preset cost function. This process It can be understood as a dynamic programming solution process in a limited space, and finally a discrete sequence that satisfies the boundary value of the longitudinal feasible range can be searched and is incremented in time, which is expressed as an s-t discrete sequence, where s represents the target vehicle corresponding to the discrete time t. the first target position.

S309. Perform a fitting process on the first target position to obtain a reference speed of the target vehicle.

In this step, after the first target position is obtained, a fitting process may be performed on the first target position to obtain the reference speed of the target vehicle. Exemplarily, by fitting the s-t discrete sequence points obtained in the example of step S308, the reference speed of the target vehicle can be obtained, and the reference speed is specifically described in the form of an s-t curve (curve).

In this embodiment of the present application, step S204 in FIG. 2 may further include the following two steps of S310 and S311:

S310. Obtain a target candidate path corresponding to a future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed.

Wherein, the preset candidate path is obtained by vertically and horizontally sampling the future travel path of the target vehicle respectively.

Exemplarily, taking the current position of the target vehicle as the longitudinal starting position, the longitudinal position is sampled at equal intervals along the direction of the reference line, and for each sampling point (represented as ss), the corresponding lane width is queried as the horizontal sampling space, The horizontal sampling space is divided at equal intervals to obtain discrete points (ss, dd), where dd represents the horizontal position of the target vehicle relative to the reference line. Using a spline curve to smoothly connect the starting point and each sampling point, multiple initial candidate paths can be generated, described in the form of ss-dd curve, that is, the preset candidate paths corresponding to the target vehicle are obtained. After the preset candidate path corresponding to the target vehicle is obtained, the target candidate path corresponding to the future preset time range may be obtained according to the preset candidate path corresponding to the target vehicle and the reference speed.

Further, optionally, obtaining the target candidate path corresponding to the future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed, including: for each preset candidate path, according to the reference speed, obtaining discrete The longitudinal advance distance corresponding to the target vehicle corresponding to the time; according to the longitudinal advance distance and the preset candidate path, the second target position of the target vehicle corresponding to the discrete time is obtained; the coordinate transformation of the second target position is performed to obtain the target candidate path.

Exemplarily, for each preset candidate path, according to equal time intervals, that is, for each discrete moment, query the reference speed curve s-t curve to obtain the longitudinal forward distance of the target vehicle corresponding to the discrete moment. s1 is represented, and the lateral position of the target vehicle relative to the reference line is obtained by s1 interpolation on the ss-dd curve. The spatial coordinates (x,y) of the target vehicle. According to the above method, a target candidate path (which may also be referred to as a target candidate trajectory) in a preset time range in the future (for example, 10s) can be deduced.

S311. Determine a target trajectory of the target vehicle within a predetermined time range in the future according to the target candidate path and the second preset cost function.

Wherein, the second preset cost function is determined based on the driving safety, comfort and stability of the target vehicle.

Exemplarily, after obtaining the target candidate path (that is, the target candidate trajectory), for each target candidate trajectory, when defining the second preset cost function (represented as cost_function_2), the potential collision risk of the target candidate trajectory needs to be comprehensively considered. , that is, the probability that the target vehicle and the obstacle appear in the same position at the same time, which is used to determine the driving safety of the target vehicle; the comfort of the trajectory, that is, the acceleration of the trajectory and its rate of change; the stability of the trajectory, that is, the target candidate trajectory Maximum yaw rate. Therefore, the definition of the second preset cost function is:

cost_function_2=coefficient_risk*trajectory_risk_value+coefficient_comfort*(trajectory_max_jerk+trajectory_max_acc)+coefficient_stability*trajectory_max_yaw_rate;

Among them, trajectory_risk_value, trajectory_max_jerk, trajectory_max_acc, and trajectory_max_yaw_rate represent the collision risk, maximum jerk, maximum acceleration, and maximum yaw angular velocity of the entire target candidate trajectory, respectively; coefficient_risk, coefficient_comfort, and coefficient_stability represent the weight factor of the potential collision risk of the target candidate trajectory, respectively, The weighting factor for comfort and the weighting factor for stability, and both are positive values.

In this step, after the target candidate path is obtained, the target trajectory of the target vehicle in the future preset time range can be determined according to the target candidate path and the second preset cost function, that is, an optimal trajectory is finally selected and released to the target The control module of the vehicle controls the target vehicle to travel according to the planned optimal trajectory.

The driving trajectory planning method provided by the embodiment of the present application determines the expected interaction decision between the target vehicle and the obstacle in the future preset time range; based on the expected interaction decision, determines the target vehicle corresponding to the discrete moment in the future preset time range Longitudinal feasible minimum distance, target longitudinal feasible maximum distance, target longitudinal velocity minimum and target longitudinal velocity maximum, according to target longitudinal feasible minimum distance, target longitudinal maximum feasible distance, target longitudinal velocity minimum and target longitudinal velocity maximum value, construct the longitudinal feasible space map of the target vehicle corresponding to the future preset time range; for the target longitudinal feasible minimum distance and the target longitudinal feasible maximum distance corresponding to discrete moments in the future preset time range, according to the target longitudinal feasible distance corresponding to the previous discrete time The minimum distance and the target longitudinal feasible minimum distance corresponding to the current discrete moment, obtain the updated target longitudinal feasible minimum distance corresponding to the current discrete moment, and obtain the current discrete target according to the target maximum feasible distance and the target longitudinal feasible maximum distance corresponding to the current discrete moment. The updated maximum longitudinal feasible distance of the target corresponding to the time; according to the updated minimum longitudinal feasible distance of the target and the updated maximum longitudinal feasible distance of the target, the updated longitudinal feasible space map is obtained; The maximum acceleration capability, the maximum deceleration capability, the curvature corresponding to the reference line of the path within the longitudinal feasible range, and the corresponding relationship between the preset curvature and the speed constraint, obtain the updated longitudinal feasible range and longitudinal speed constraint information; according to the updated longitudinal feasible range range and longitudinal speed constraint information to obtain the updated longitudinal feasible space map; based on the updated longitudinal feasible space map and the first preset cost function, obtain the first target position of the target vehicle corresponding to the discrete time; Perform a fitting process to obtain the reference speed of the target vehicle; obtain the target candidate path corresponding to the future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed; according to the target candidate path and the second preset cost function , to determine the target trajectory of the target vehicle within a preset time range in the future. Since the embodiment of the present application fully considers the information of static obstacles and dynamic obstacles in the driving scene, the desired interaction decision is determined, and then the longitudinal feasible space map of the target vehicle is constructed. Based on the longitudinal feasible space map and the execution capability of the target vehicle, the Under the premise of reasonably interacting with obstacles, the reference speed closest to the final execution of the target vehicle is obtained, and the reference speed is used to evaluate the optimal path. Therefore, the optimal path can be obtained more accurately, and the reasonableness of the driving trajectory planning result can be improved. performance and reliability, improve driving efficiency and improve driving experience.

The following are apparatus embodiments of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

FIG. 4 is a schematic structural diagram of a driving trajectory planning device according to an embodiment of the present application. As shown in FIG. 4 , the driving trajectory planning device 400 in the embodiment of the present application includes: a determination module 401 , a construction module 402 , an acquisition module 403 and a processing module 404. in:

The determination module 401 is used to determine the expected interaction decision between the target vehicle and the obstacle in the future preset time range. The expected interaction decision is used to characterize the interaction behavior of the target vehicle and the obstacle and the interaction time window corresponding to the interaction behavior. The interaction behavior includes The target vehicle actively overtakes the obstacle, the target vehicle actively yields to the obstacle, and the target vehicle actively ignores the obstacle.

The building module 402 is used for constructing a longitudinal feasible space map of the target vehicle corresponding to a preset time range in the future based on the expected interaction decision, and the longitudinal feasible space map is used to represent the longitudinal feasible range of the target vehicle corresponding to discrete moments in the preset time range in the future and Longitudinal velocity constraint information.

The obtaining module 403 is configured to obtain the reference speed of the target vehicle based on the longitudinal feasible space map and a first preset cost function, where the first preset cost function is determined based on the driving efficiency and comfort of the target vehicle.

The processing module 404 is configured to determine, based on the reference speed, a target trajectory of the target vehicle within a preset time range in the future.

In some embodiments, the building module 402 may be specifically configured to: determine, based on the desired interaction decision, the minimum target longitudinal feasible distance, the target maximum feasible longitudinal distance, and the minimum target longitudinal speed corresponding to the target vehicle at discrete moments in the future preset time range. value and the maximum value of the target longitudinal speed; according to the minimum feasible longitudinal distance of the target, the maximum feasible longitudinal distance of the target, the minimum value of the target longitudinal speed and the maximum value of the target longitudinal speed, construct the longitudinal feasible space map of the target vehicle corresponding to the preset time range in the future .

Optionally, the building module 402 is used to make decisions based on the expected interaction and determine the target longitudinal minimum feasible distance, the target longitudinal feasible maximum distance, the target longitudinal velocity minimum, and the target longitudinal distance corresponding to the target vehicle at discrete moments in the future preset time range. The maximum value of the speed can be specifically used to: determine the initial maximum feasible longitudinal distance of the target vehicle, the minimum initial feasible longitudinal distance, the minimum initial longitudinal speed, and the maximum initial longitudinal speed; for discrete moments within a preset time range in the future For each obstacle corresponding to it, perform the following operations until each obstacle is traversed: Based on the expected interaction decision, if it is determined that the target vehicle needs to actively yield to the target obstacle, according to the space occupied by the target obstacle longitudinally and laterally, Assuming the yield line safety distance and the initial maximum feasible longitudinal distance, determine the target maximum longitudinal feasible distance of the target vehicle, and determine the maximum longitudinal speed of the target according to the speed of the target obstacle, and determine the maximum feasible longitudinal distance of the target as the new initial longitudinal feasible distance. Maximum distance, determine the maximum value of the target longitudinal speed as the maximum value of the new initial longitudinal speed; or, based on the expected interaction decision, if it is determined that the target vehicle needs to actively surpass the target obstacle, the space occupied by the target obstacle longitudinally and laterally is determined according to the , preset the overtaking safety distance and the initial minimum feasible longitudinal distance, determine the target minimum longitudinal feasible distance of the target vehicle, and determine the minimum value of the target longitudinal speed according to the speed of the target obstacle, and determine the target longitudinal minimum feasible distance as the new initial longitudinal The minimum feasible distance is determined, and the minimum value of the target longitudinal velocity is determined as the minimum value of the new initial longitudinal velocity.

Optionally, when the building module 402 is used to determine the initial longitudinal feasible maximum distance of the target vehicle, it may be specifically used to: determine the maximum feasible distance of the target vehicle within a preset time range in the future based on the speed limit of the driving scene of the target vehicle; Determine the maximum speed corresponding to the maximum curvature according to the maximum curvature of the path traveled by the target vehicle within the preset time range in the future and the corresponding relationship between the preset curvature and the speed constraint; obtain the maximum feasible distance of the target according to the maximum speed and the maximum feasible distance; determine The target maximum feasible distance is the initial vertical maximum feasible distance.

In some embodiments, the driving trajectory planning device further includes an update module 405 for, after the building module 402 builds the longitudinal feasible space map of the target vehicle corresponding to the future preset time range, for the future preset time range corresponding to discrete moments The target longitudinal feasible minimum distance and the target longitudinal maximum feasible distance are obtained according to the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment, and the updated target longitudinal feasible minimum distance corresponding to the current discrete moment is obtained. , and obtain the updated maximum longitudinal feasible distance of the target corresponding to the current discrete moment according to the maximum feasible distance of the target and the maximum longitudinal feasible distance of the target corresponding to the current discrete moment; The maximum distance to obtain the updated vertical feasible space map.

Optionally, the update module 405 is used to obtain the updated target longitudinal feasible minimum distance corresponding to the current discrete moment according to the target longitudinal feasible minimum distance corresponding to the previous discrete moment and the target longitudinal feasible minimum distance corresponding to the current discrete moment, and According to the maximum feasible distance of the target and the maximum vertical feasible distance of the target corresponding to the current discrete time, when obtaining the updated maximum vertical feasible distance of the target corresponding to the current discrete time, it can be specifically used for: for the discrete time within the preset time range in the future, if it is determined If the updated maximum longitudinal feasible distance of the target is smaller than the updated minimum longitudinal feasible distance of the target, the interaction behavior between the corresponding obstacle and the target vehicle is updated, and the longitudinal feasible space map of the target vehicle corresponding to the preset time range in the future is reconstructed.

In some embodiments, the obtaining module 403 may be specifically configured to: based on the longitudinal feasible space map, according to the maximum acceleration capability, the maximum deceleration capability of the target vehicle, the curvature corresponding to the reference line of the path within the longitudinal feasible range, and the preset curvature and speed The corresponding relationship of constraints, obtain the updated longitudinal feasible range and longitudinal speed constraint information; according to the updated longitudinal feasible range and longitudinal speed constraint information, obtain the updated longitudinal feasible space map; based on the updated longitudinal feasible space map and the first A preset cost function is used to obtain the first target position of the target vehicle corresponding to discrete moments; the first target position is fitted to obtain the reference speed of the target vehicle.

In some embodiments, the processing module 404 may be specifically configured to: obtain a target candidate path corresponding to a preset time range in the future according to at least one preset candidate path corresponding to the target vehicle and the reference speed, and the preset candidate path is a future target vehicle for the target vehicle. The driving path is obtained by vertical and horizontal sampling respectively; according to the target candidate path and the second preset cost function, the target trajectory of the target vehicle in the future preset time range is determined, and the second preset cost function is based on the driving safety of the target vehicle. degree, comfort and stability.

Optionally, when the processing module 404 is used to obtain the target candidate path corresponding to the future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed, it can be specifically used for: for each preset candidate path. , according to the reference speed, the longitudinal advance distance corresponding to the target vehicle corresponding to the discrete time is obtained; according to the longitudinal advance distance and the preset candidate path, the second target position of the target vehicle corresponding to the discrete time is obtained; the coordinate transformation is performed on the second target position, Get the target candidate path.

The apparatus in this embodiment can be used to implement the technical solutions of any of the above-described method embodiments, and the implementation principles and technical effects thereof are similar, and are not repeated here.

FIG. 5 is a schematic structural diagram of an electronic device provided by the present application. As shown in FIG. 5 , the electronic device 500 may include: at least one processor 501 and a memory 502 .

The memory 502 is used to store programs. Specifically, the program may include program code, and the program code includes computer operation instructions.

The memory 502 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

The processor 501 is configured to execute the computer-executed instructions stored in the memory 502, so as to implement the driving trajectory planning method described in the foregoing method embodiments. The processor 501 may be a central processing unit (Central Processing Unit, referred to as CPU), or a specific integrated circuit (Application Specific Integrated Circuit, referred to as ASIC), or is configured to implement one or more of the embodiments of the present application an integrated circuit. Specifically, when implementing the driving trajectory planning method described in the foregoing method embodiments, the electronic device may be, for example, an electronic device with a processing function such as a terminal and a server. When implementing the driving trajectory planning method described in the foregoing method embodiments, the electronic device may be, for example, an electronic control unit on a vehicle.

Optionally, the electronic device 500 may further include a communication interface 503 . In terms of specific implementation, if the communication interface 503 , the memory 502 and the processor 501 are implemented independently, the communication interface 503 , the memory 502 and the processor 501 can be connected to each other through a bus and communicate with each other. The bus may be an industry standard architecture (Industry Standard Architecture, ISA) bus, a Peripheral Component (Peripheral Component, PCI) bus, or an Extended Industry Standard Architecture (Extended Industry Standard Architecture, EISA) bus, or the like. Buses can be divided into address bus, data bus, control bus, etc., but it does not mean that there is only one bus or one type of bus.

Optionally, in terms of specific implementation, if the communication interface 503 , the memory 502 and the processor 501 are integrated on one chip, the communication interface 503 , the memory 502 and the processor 501 can communicate through an internal interface.

The application also provides a computer-readable storage medium, the computer-readable storage medium may include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory) Various media that can store program codes, such as a magnetic disk, a magnetic disk, or an optical disk, specifically, the computer-readable storage medium stores program instructions, and the program instructions are used in the driving trajectory planning method in the above embodiment.

The present application also provides a computer program product comprising execution instructions stored in a readable storage medium. At least one processor of the electronic device may read the execution instruction from the readable storage medium, and the execution of the execution instruction by the at least one processor causes the electronic device to implement the driving trajectory planning method provided by the above-mentioned various embodiments.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features thereof can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present application. scope.

Claims (13)

1. A method for planning a driving trajectory is characterized by comprising the following steps:

determining a desired interaction decision of a target vehicle and an obstacle within a future preset time range, wherein the desired interaction decision is used for representing an interaction behavior of the target vehicle and the obstacle and an interaction time window corresponding to the interaction behavior, and the interaction behavior comprises that the target vehicle actively exceeds the obstacle, the target vehicle actively yields the obstacle and the target vehicle actively ignores the obstacle;

constructing a longitudinal feasible space map of the target vehicle corresponding to the future preset time range based on the expected interaction decision, wherein the longitudinal feasible space map is used for representing a longitudinal feasible range and longitudinal speed constraint information of the target vehicle corresponding to discrete time within the future preset time range;

obtaining a reference speed of the target vehicle based on the longitudinal feasible space diagram and a first preset cost function, wherein the first preset cost function is determined based on the running efficiency and the comfort degree of the target vehicle;

determining a target trajectory of the target vehicle within the future preset time range based on the reference speed.

2. The method for planning a driving trajectory according to claim 1, wherein the constructing a longitudinal feasible space map of the target vehicle corresponding to the future preset time range based on the desired interaction decision comprises:

determining a target longitudinal feasible minimum distance, a target longitudinal feasible maximum distance, a target longitudinal speed minimum value and a target longitudinal speed maximum value corresponding to discrete time of the target vehicle in the future preset time range based on the expected interaction decision;

and constructing a longitudinal feasible space map of the target vehicle corresponding to the future preset time range according to the target longitudinal feasible minimum distance, the target longitudinal feasible maximum distance, the minimum value of the target longitudinal speed and the maximum value of the target longitudinal speed.

3. The method for planning a driving trajectory according to claim 2, wherein the determining, based on the desired interaction decision, a minimum target longitudinal feasible distance, a maximum target longitudinal feasible distance, a minimum target longitudinal speed, and a maximum target longitudinal speed corresponding to the target vehicle at the discrete time within the future preset time range comprises:

determining an initial longitudinally feasible maximum distance, an initial longitudinally feasible minimum distance, a minimum value of initial longitudinal velocity, and a maximum value of initial longitudinal velocity for the target vehicle;

and executing the following operations for each obstacle corresponding to the discrete time within the future preset time range until each obstacle is traversed:

based on the expected interactive decision, if it is determined that the target vehicle is required to actively yield the target obstacle, determining a target longitudinal feasible maximum distance of the target vehicle according to a space occupied by the target obstacle in the longitudinal and transverse directions, a preset yielding safety distance and the initial longitudinal feasible maximum distance, determining a maximum value of a target longitudinal speed according to the speed of the target obstacle, determining the target longitudinal feasible maximum distance as a new initial longitudinal feasible maximum distance, and determining the maximum value of the target longitudinal speed as a new maximum value of an initial longitudinal speed;

or, based on the expected interactive decision, if it is determined that the target vehicle is required to actively exceed the target obstacle, determining a target longitudinal feasible minimum distance of the target vehicle according to a space occupied by the target obstacle in the longitudinal and lateral directions, a preset overtaking safety distance and the initial longitudinal feasible minimum distance, determining a minimum value of a target longitudinal speed according to the speed of the target obstacle, determining the target longitudinal feasible minimum distance as a new initial longitudinal feasible minimum distance, and determining the minimum value of the target longitudinal speed as a new initial longitudinal speed minimum value.

4. The method of claim 3, wherein said determining an initial longitudinal feasible maximum distance of the target vehicle comprises:

determining the maximum feasible distance of the target vehicle within the future preset time range based on the driving scene speed limit of the target vehicle;

determining the maximum speed corresponding to the maximum curvature according to the maximum curvature of the running path of the target vehicle in the future preset time range and the corresponding relation between the preset curvature and the speed constraint;

obtaining a target maximum feasible distance according to the maximum speed and the maximum feasible distance;

determining the target maximum feasible distance as the initial longitudinally feasible maximum distance.

5. The method for planning a driving trajectory according to claim 4, wherein after the constructing the longitudinal feasible space map of the target vehicle corresponding to the preset time range in the future, the method further comprises:

aiming at the target longitudinal feasible minimum distance and the target longitudinal feasible maximum distance corresponding to the discrete time within the future preset time range, obtaining the updated target longitudinal feasible minimum distance corresponding to the current discrete time according to the target longitudinal feasible minimum distance corresponding to the previous discrete time and the target longitudinal feasible minimum distance corresponding to the current discrete time, and obtaining the updated target longitudinal feasible maximum distance corresponding to the current discrete time according to the target maximum feasible distance and the target longitudinal feasible maximum distance corresponding to the current discrete time;

and obtaining an updated longitudinal feasible space map according to the updated minimum distance of the target longitudinal feasible and the updated maximum distance of the target longitudinal feasible.

6. The method for planning a driving trajectory according to claim 5, wherein the obtaining an updated target longitudinally feasible minimum distance corresponding to the current discrete time according to a target longitudinally feasible minimum distance corresponding to a previous discrete time and a target longitudinally feasible minimum distance corresponding to the current discrete time, and obtaining an updated target longitudinally feasible maximum distance corresponding to the current discrete time according to the target maximum feasible distance and a target longitudinally feasible maximum distance corresponding to the current discrete time comprise:

and if the updated maximum distance of the target longitudinal feasible is determined to be smaller than the updated minimum distance of the target longitudinal feasible for the discrete time within the future preset time range, updating the interaction behavior of the corresponding barrier and the target vehicle, and reconstructing a longitudinal feasible space diagram of the target vehicle corresponding to the future preset time range.

7. The driving trajectory planning method according to any one of claims 1 to 6, wherein the obtaining the reference speed of the target vehicle based on the longitudinal feasible space map and a first preset cost function comprises:

based on the longitudinal feasible space map, obtaining updated longitudinal feasible range and longitudinal speed constraint information according to the maximum acceleration capacity and the maximum deceleration capacity of the target vehicle, the curvature corresponding to the reference line of the path in the longitudinal feasible range and the corresponding relation between preset curvature and speed constraint;

obtaining an updated longitudinal feasible space map according to the updated longitudinal feasible range and longitudinal speed constraint information;

acquiring a first target position of the target vehicle corresponding to the discrete moment based on the updated longitudinal feasible space diagram and the first preset cost function;

and fitting the first target position to obtain the reference speed of the target vehicle.

8. The method of claim 7, wherein the determining the target trajectory of the target vehicle within the future preset time range based on the reference speed comprises:

obtaining a target candidate path corresponding to the future preset time range according to at least one preset candidate path corresponding to the target vehicle and the reference speed, wherein the preset candidate path is obtained by longitudinally and transversely sampling the future running path of the target vehicle;

and determining a target track of the target vehicle in the future preset time range according to the target candidate path and a second preset cost function, wherein the second preset cost function is determined based on the driving safety degree, the comfort degree and the stability of the target vehicle.

9. The method for planning a driving trajectory according to claim 8, wherein the obtaining a target candidate route corresponding to the future preset time range according to at least one preset candidate route corresponding to the target vehicle and the reference speed comprises:

for each preset candidate path, obtaining a longitudinal advance distance corresponding to the target vehicle corresponding to the discrete moment according to the reference speed;

according to the longitudinal advancing distance and the preset candidate path, a second target position of the target vehicle corresponding to the discrete moment is obtained;

and carrying out coordinate transformation on the second target position to obtain the target candidate path.

10. A trajectory planning device, comprising:

the system comprises a determining module, a judging module and a judging module, wherein the determining module is used for determining an expected interaction decision of a target vehicle and an obstacle in a future preset time range, the expected interaction decision is used for representing an interaction behavior of the target vehicle and the obstacle and an interaction time window corresponding to the interaction behavior, and the interaction behavior comprises that the target vehicle actively exceeds the obstacle, the target vehicle actively gives way to the obstacle and the target vehicle actively ignores the obstacle;

the construction module is used for constructing a longitudinal feasible space map of the target vehicle corresponding to the future preset time range based on the expected interaction decision, wherein the longitudinal feasible space map is used for representing the longitudinal feasible range and longitudinal speed constraint information of the target vehicle corresponding to discrete time within the future preset time range;

an obtaining module, configured to obtain a reference speed of the target vehicle based on the longitudinal feasible space diagram and a first preset cost function, where the first preset cost function is determined based on a driving efficiency and a comfort level of the target vehicle;

and the processing module is used for determining a target track of the target vehicle in the future preset time range based on the reference speed.

11. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;

the memory stores computer-executable instructions;

the processor executes computer-executable instructions stored by the memory to implement the trajectory planning method of any one of claims 1 to 9.

12. A computer-readable storage medium, in which computer program instructions are stored, which, when executed by a processor, implement a driving trajectory planning method according to any one of claims 1 to 9.

13. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements a driving trajectory planning method according to any one of claims 1 to 9.

Images