Vehicle trajectory planning method and device, unmanned vehicle and storage medium

Abstract

The present application discloses a vehicle trajectory planning method and device, an unmanned vehicle and a storage medium, wherein the method includes obtaining the driving environment information of the target vehicle, wherein the driving environment information includes at least one of road boundary information, lane line information, obstacle information and target position information; calculating the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle; constructing a danger avoidance corridor accessible to the target vehicle based on the obstacle avoidance direction and obstacle information; and planning the driving trajectory of the target vehicle based on the danger avoidance corridor. Such a trajectory planning method provides the obstacle avoidance breakthrough direction of the vehicle through an adaptive potential field to open a breakthrough at a reasonable position on the lane boundary, thereby ensuring the stability of the vehicle trajectory planning decision.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology and specifically relates to a vehicle trajectory planning method and device, an unmanned vehicle and a storage medium. Background Art

[0002] In the scenario of low-speed delivery by autonomous driving, when a nearby vehicle suddenly changes lanes / cuts in, and the vehicle cannot avoid a collision by emergency braking along the current lane at the current speed, it will need to turn and brake to avoid a collision. However, if the lane boundary in the turning direction of the current lane is a strong constraint, the vehicle will face a dilemma, resulting in indecisive or unstable vehicle behavior, or even a collision.

[0003] The information disclosed in this background technology section is only intended to increase the understanding of the overall background of the application and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0004] The purpose of the present application is to provide a vehicle trajectory planning method, which is used to solve the problem of how to avoid obstacles that cut into the lane.

[0005] To achieve the above objectives, the present application provides a vehicle trajectory planning method, the method comprising:

[0006] Acquiring driving environment information of the target vehicle, wherein the driving environment information includes at least one of road boundary information, lane line information, obstacle information, and target position information;

[0007] Calculating the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle;

[0008] Based on the obstacle avoidance direction and obstacle information, constructing a danger avoidance corridor passable by the target vehicle;

[0009] Based on the escape corridor, a driving trajectory of the target vehicle is planned.

[0010] In one embodiment, based on the obstacle avoidance direction and obstacle information, constructing a traversable corridor for the target vehicle specifically includes:

[0011] Based on the obstacle information, determining an inaccessible area for the target vehicle;

[0012] Determine a first boundary based on a road boundary or a lane line pointed to by the obstacle avoidance direction, wherein the first boundary and the impassable area do not interfere with each other;

[0013] Based on the first boundary and the impassable area, a safe haven corridor for the target vehicle is constructed.

[0014] In one embodiment, based on the obstacle avoidance direction and obstacle information, constructing a traversable corridor for the target vehicle specifically includes:

[0015] Determining a first reference point based on a position of the impassable area closest to the first boundary;

[0016] Based on the longitudinal position of the target vehicle intruding into the obstacle avoidance lane in the obstacle avoidance direction, determining a corresponding second reference point on the first boundary, wherein the obstacle avoidance lane is a lane corresponding to the first boundary;

[0017] Based on the first reference point, determining a first corridor boundary, wherein the first corridor boundary includes a first boundary segment longitudinally adjacent to the target vehicle and a second boundary segment longitudinally away from the target vehicle, the first boundary segment starts from the first reference point and is parallel to the obstacle avoidance direction, and the second boundary segment starts from the first reference point and is parallel to the first boundary;

[0018] Based on the second reference point, determining a second corridor boundary, wherein the second corridor boundary includes a third boundary segment longitudinally adjacent to the target vehicle and a fourth boundary segment longitudinally away from the target vehicle, the third boundary segment takes the second reference point as a starting point and is parallel to the obstacle avoidance direction, and the fourth boundary segment takes the second reference point as a starting point and is parallel to the first boundary;

[0019] A danger avoidance corridor for the target vehicle is constructed based on the first corridor boundary and the second corridor boundary.

[0020] In one embodiment, determining the impassable area of ​​the target vehicle based on the obstacle information specifically includes:

[0021] Based on the maximum steering angle and driving speed of the obstacle, the impassable area of ​​the target vehicle is determined, wherein the impassable area of ​​the target vehicle is an area that can be covered by the obstacle within a preset future time period.

[0022] In one embodiment, planning the driving trajectory of the target vehicle based on the danger avoidance corridor specifically includes:

[0023] Determine a plurality of corresponding driving trajectories based on a plurality of preset semantic action sequence templates, wherein the semantic action sequence templates include semantic action directions within a plurality of planning cycles, and the semantic action directions include any one of a right turn, a left turn, and action holding;

[0024] The driving trajectory of the target vehicle is determined among the multiple driving trajectories with the goal of maximizing a preset reward function, wherein the preset reward function is constructed based on a risk avoidance corridor.

[0025] In one embodiment, planning the driving trajectory of the target vehicle based on the danger avoidance corridor further includes:

[0026] Determining a decision direction corresponding to the obstacle avoidance direction, wherein the decision direction includes a right turn and a left turn;

[0027] When the decision direction is a right turn, multiple preset first-category semantic action sequence templates are determined, and when the decision direction is a left turn, multiple preset second-category semantic action sequence templates are determined, wherein the semantic action directions corresponding to the first-category semantic action sequence templates include right turn, or right turn and action maintenance, and the semantic action directions corresponding to the second-category semantic action sequence templates include left turn, or left turn and action maintenance.

[0028] In one embodiment, based on a plurality of preset semantic action sequence templates, a plurality of corresponding driving trajectories are determined, specifically including:

[0029] Determining a front wheel turning angle of the target vehicle in the current planning cycle based on a preview point of the target vehicle in the danger avoidance corridor in the current planning cycle;

[0030] Based on the obstacle avoidance direction of the target vehicle, determine the lateral obstacle information in front of the target vehicle, wherein the lateral obstacle information includes speed information and longitudinal distance information from the target vehicle;

[0031] Determining the longitudinal speed of the target vehicle in a current planning cycle based on the lateral obstacle information;

[0032] Based on the front wheel turning angle and the longitudinal speed of the target vehicle in the current planning period, the trajectory state information of the target vehicle in the current planning period is determined.

[0033] In one embodiment, the front wheel turning angle of the target vehicle in the current planning cycle is constrained by the front wheel turning angle acceleration, and / or the front wheel turning angle jump, and / or the maximum front wheel turning angle velocity, and / or the maximum front wheel turning angle;

[0034] and / or,

[0035] The longitudinal speed of the target vehicle in the current planning cycle is constrained by the longitudinal acceleration, and / or the longitudinal jump, and / or the non-reverse direction.

[0036] In one embodiment, the preset reward function includes at least one of a safety distance cost sub-item, a safety speed cost sub-item, and a lateral and longitudinal coordination cost sub-item, wherein the safety distance cost sub-item represents the offset cost of the driving trajectory in each planning cycle relative to the reference trajectory in the risk avoidance corridor, the safety speed cost sub-item represents the longitudinal speed cost of the driving trajectory in each planning cycle, and the lateral and longitudinal coordination cost sub-item represents the difference cost of the longitudinal speed of the target vehicle in each planning cycle relative to the reference longitudinal speed, and the reference longitudinal speed is the longitudinal speed corresponding to the smaller of the rollover torque and the sideslip torque of the target vehicle.

[0037] In one embodiment, the method further comprises:

[0038] The desired lateral boundary is determined by shrinking the obstacle avoidance lane away from the boundary of the target vehicle by a preset distance, wherein the obstacle avoidance lane is a lane corresponding to a first boundary, the first boundary is a road boundary or lane line pointed by the obstacle avoidance direction, and the first boundary and the impassable area of ​​the target vehicle do not interfere with each other;

[0039] Determine the trajectory of the target vehicle pointing to the desired lateral boundary in the obstacle avoidance direction as a first reference trajectory segment;

[0040] Determine a segment of the desired lateral boundary that is longitudinally away from a reference position as a second reference trajectory segment, wherein the reference position is an intersection position of the desired lateral boundary and the first reference trajectory segment;

[0041] A reference trajectory in the hazard avoidance corridor is determined based on the first reference trajectory segment and the second reference trajectory segment.

[0042] In one embodiment, calculating the potential field of the target vehicle based on the driving environment information specifically includes:

[0043] Based on the road boundary information, calculating the repulsive potential field of the road boundary on the target vehicle;

[0044] Based on the lane line information, calculating the repulsive potential field of the lane line on the target vehicle;

[0045] Based on the obstacle information, calculating the repulsive potential field of the obstacle on the target vehicle;

[0046] Based on the target position information, calculating the gravitational potential field of the target position on the target vehicle;

[0047] The resultant potential field is calculated based on the repulsive potential field of the road boundary on the target vehicle, and / or the repulsive potential field of the lane line on the target vehicle, and / or the repulsive potential field of the obstacle on the target vehicle, and / or the gravitational potential field of the target position on the target vehicle to determine the potential field of the target vehicle.

[0048] In one embodiment, the method further comprises:

[0049] Obtaining the longitudinal speeds of the target vehicle and the obstacle, and the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle;

[0050] Determining a braking distance of the target vehicle based on the longitudinal speed information and the braking response delay of the target vehicle;

[0051] Determining a deceleration distance of the obstacle based on the longitudinal speed information and the risk deceleration of the obstacle;

[0052] Based on the relative distance between the obstacle and the target vehicle in the driving direction of the target vehicle, the braking distance of the target vehicle and the deceleration distance of the obstacle, it is determined whether the target vehicle avoids the obstacle.

[0053] The present application also provides a vehicle trajectory planning device, comprising:

[0054] An acquisition module, used to acquire driving environment information of a target vehicle, wherein the driving environment information includes at least one of road boundary information, lane line information, obstacle information and target position information;

[0055] A calculation module, used to calculate the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle;

[0056] A construction module, used to construct a danger avoidance corridor passable by the target vehicle based on the obstacle avoidance direction and obstacle information;

[0057] A planning module is used to plan the driving trajectory of the target vehicle based on the escape corridor.

[0058] The present application also provides an unmanned vehicle, comprising:

[0059] at least one processor; and

[0060] A memory storing instructions, which, when executed by the at least one processor, enables the at least one processor to execute the vehicle trajectory planning method as described above.

[0061] The present application also provides a machine-readable storage medium storing executable instructions, which, when executed, enable the machine to execute the vehicle trajectory planning method as described above.

[0062] Compared with the prior art, the vehicle trajectory planning method of the present application calculates the potential field of the target vehicle through the driving environment information to determine its obstacle avoidance direction, and then constructs a risk avoidance corridor based on the obstacle avoidance direction and obstacle information for planning the driving trajectory of the target vehicle. In this way, the obstacle avoidance breakthrough direction of the vehicle is provided through the adaptive potential field to open a breakthrough (break through the strong constraint) at a reasonable position on the lane boundary, thereby ensuring the stability of the vehicle trajectory planning decision.

[0063] On the other hand, when planning the target vehicle's driving trajectory in the escape corridor, the driving trajectory is determined based on the deduction method using the semantic action sequence template, and combined with the kinematic constraints to ensure the executability of the driving trajectory and reduce the requirements for computing resources for trajectory solving. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is an application scenario diagram of a vehicle trajectory planning method according to an embodiment of the present application;

[0065] Figure 2 is a flow chart of a vehicle trajectory planning method according to an embodiment of the present application;

[0066] Figure 3 and Figure 4 A scene diagram of a danger avoidance corridor is constructed according to a vehicle trajectory planning method in different embodiments of the present application;

[0067] Figure 5 is a schematic diagram of a semantic action sequence template in a vehicle trajectory planning method according to an embodiment of the present application;

[0068] Figure 6 It is a schematic diagram of selecting a corresponding type of semantic action sequence template based on a decision direction in a vehicle trajectory planning method according to an embodiment of the present application;

[0069] Figure 7 It is a flowchart of a method for deducing a driving trajectory based on a preset semantic action sequence template in a vehicle trajectory planning method according to an embodiment of the present application;

[0070] Figure 8 A scene graph for determining a reference trajectory in a vehicle trajectory planning method according to an embodiment of the present application;

[0071] Fig. 9 A scene diagram for determining whether there is a risk of collision between a target vehicle and an intervening obstacle in a vehicle trajectory planning method according to an embodiment of the present application;

[0072] Fig.10 A module diagram of a vehicle trajectory planning device according to an embodiment of the present application;

[0073] Fig.11It is a hardware structure diagram of an unmanned vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0074] The present application will be described in detail below in conjunction with the various embodiments shown in the accompanying drawings. However, these embodiments do not limit the present application, and any structural, methodological, or functional changes made by a person of ordinary skill in the art based on these embodiments are all within the scope of protection of the present application.

[0075] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "corresponding to" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0076] Before introducing the embodiments of the present application, the basic technologies and some technical terms involved in the embodiments of the present application are schematically explained:

[0077] Autonomous driving system: A system that implements different levels of autonomous driving functions of a vehicle, such as assisted driving system (L2), high-speed autonomous driving system that requires human supervision (L3), and highly / fully autonomous driving system (L4 / L5).

[0078] Intelligent Traffic System (ITS), also known as Intelligent Transportation System, effectively and comprehensively applies advanced science and technology (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operations research and artificial intelligence, etc.) to transportation, service control and vehicle manufacturing, strengthens the connection between vehicles, roads and users, and thus forms a comprehensive transportation system that ensures safety, improves efficiency, improves the environment and saves energy.

[0079] Intelligent Vehicle Infrastructure Cooperative Systems (IVICS), referred to as IVICS, is a development direction of intelligent transportation systems. IVICS uses advanced wireless communications and new generation Internet technologies to implement dynamic real-time information interaction between vehicles and between vehicles and roads in all directions, and conducts active vehicle safety control and road cooperative management based on the collection and integration of dynamic traffic information in all time and space, fully realizing the effective coordination of people, vehicles and roads, ensuring traffic safety, and improving traffic efficiency, thus forming a safe, efficient and environmentally friendly road traffic system.

[0080] Trajectory planning refers to the initial state of a given vehicle (including starting position, speed and acceleration), target state (including target position, speed and acceleration), obstacle position, dynamics and comfort. A smooth trajectory is calculated based on the constraints, and the vehicle is guided along this trajectory to reach the target state. Trajectory planning consists of two parts: path planning and speed planning. Path planning is responsible for calculating a smooth path from the starting position to the target position, while speed planning calculates the speed of each path point based on this path to form a speed curve.

[0081] Frenet coordinate system: also called road coordinate system, with the vehicle's starting position as the origin. The coordinate axes are perpendicular to each other and are divided into the S-axis direction (i.e. the tangent direction along the road reference line, called the longitudinal direction) and the L-axis direction (i.e. the current normal direction of the reference line, called the transverse direction). The coordinates are expressed as (S, L).

[0082] Figure 1 This is an optional schematic diagram of the system architecture involved in the embodiment of the present application. Figure 1 As shown in the figure, sensors (such as laser radar, visual sensors, etc.) collect information about the target vehicle's driving environment, and the sensors send the collected information to the server. After receiving the information, the server calculates the potential field of the vehicle and determines the obstacle avoidance direction of the vehicle for obstacles in the driving environment. The server also constructs a safe corridor for vehicles to pass based on the obstacle avoidance direction and obstacle information in the driving environment, and plans the driving trajectory based on the safe corridor.

[0083] Continue to participate Figure 1, in a specific scenario of applying the vehicle trajectory planning method of the present application, a server and a terminal device may be included. Among them, the terminal device may include a vehicle-mounted terminal and a user terminal. The vehicle-mounted terminal may include a driving computer or an on-board unit (On Board Unit, OBU), etc. The vehicle-mounted terminal may also be an application (application, APP) on the terminal, an APP on the smart rearview mirror, an APP or a small program on a mobile phone, etc., which are not limited here. The user terminal (userequipment, UE) can be a wireless terminal device or a wired terminal device. The wireless terminal device can refer to a device with a wireless transceiver function. The user terminal can be a mobile phone (mobile phone), a tablet computer (Pad), a computer with a wireless transceiver function, a virtual reality (Virtual Reality, VR) user device, an augmented reality (Augmented Reality, AR) user device, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, etc., which are not limited here. The terminal communicates with the server through the network. The data storage system can store the data that the server needs to process. The data storage system can exist alone or be integrated on the server.

[0084] It can be understood that the above is just an example of a possible execution subject of the vehicle trajectory planning method of the present application, which is explained by way of example using a server. In more embodiments, the vehicle trajectory planning can also be directly executed by a terminal device with sufficient computing power, and the present application does not limit this. Moreover, regardless of the type of server / terminal device, the vehicle trajectory planning method provided in the embodiment of the present application can be adapted to the autonomous driving system of unmanned vehicles, including L2, L3, L4 and above autonomous driving systems.

[0085] Specific reference Figure 2 , introduces an embodiment of the vehicle trajectory planning method of the present application. In this embodiment, the method includes:

[0086] S11. Acquire the driving environment information of the target vehicle.

[0087] The driving environment information of the target vehicle may include at least one of road boundary information, lane line information, obstacle information, and target position information. Exemplarily, the road boundary information may include the lateral coordinate position of the road boundary, the lane line information may include the number of lane lines and the lateral coordinate position of the corresponding lane lines, and the obstacle information may include the number of obstacles and the corresponding distance between each obstacle and the target vehicle.

[0088] In an embodiment of the present application, road boundaries may include double yellow lines used to distinguish the direction of vehicle travel, actual physical boundaries of the road, etc.; obstacles may include stationary obstacles in the road, moving obstacles (such as vehicles) cutting into the lane, etc., and the present application does not impose any restrictions on this.

[0089] S12. Calculate the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle.

[0090] Specifically, based on the road boundary information, the repulsive potential field of the road boundary on the target vehicle can be calculated, which is expressed as follows:

[0091] (1)

[0092] Where L represents the current lateral coordinate position of the target vehicle, L r,i Indicates the horizontal coordinate position of the left / right boundary of the road. Determines the rate at which the repulsive potential field of the road boundary on the target vehicle rises or falls.

[0093] Based on the lane line information, the repulsive potential field of the lane line on the target vehicle can be calculated, which is expressed as follows:

[0094] (2)

[0095] Where N is the number of lanes, α lane represents the maximum amplitude of the repulsive potential field of the lane line on the target vehicle, L represents the current lateral coordinate position of the target vehicle, and L l,i represents the lateral coordinate position of the i-th lane, Determines the rate at which the repulsive potential field of the lane line on the target vehicle rises or falls.

[0096] Based on the obstacle information, the repulsive potential field of the obstacle on the target vehicle can be calculated, which is expressed as follows:

[0097]

[0098] Among them, α obs Represents the maximum amplitude of the repulsive potential field of the obstacle on the target vehicle, p=(s,l) T is the coordinate of the target vehicle in the Frenet coordinate system.

[0099] For example, assuming there are M obstacles, the repulsive potential field of these M obstacles on the target vehicle is expressed as follows:

[0100]

[0101] in, and are the repulsive potential fields of the obstacle on the target vehicle in the longitudinal and lateral directions respectively. The repulsive potential fields of the obstacle on the target vehicle in the longitudinal and lateral directions obey Gaussian distribution with means μ1 and μ2 and covariances Σ1 and Σ2.

[0102] Based on the target position information, the gravitational potential field of the target position on the target vehicle can be calculated, which is expressed as follows:

[0103] (5)

[0104] Where p=(s,l) T is the coordinate of the target vehicle in the Frenet coordinate system, ptar=(star,ltar) T is the coordinate of the target position in the Frenet coordinate system.

[0105] Among them, the repulsive potential field of the road boundary on the target vehicle and the repulsive potential field of the lane line on the target vehicle can be used to prevent the vehicle from driving off the road and guide the target vehicle to stay in the middle of the road; the repulsive potential field of the obstacle on the target vehicle can keep a suitable safe distance between the target vehicle and the obstacle, and the target vehicle can change lanes or operate the brake lights according to the changes in the potential field; the gravitational potential field of the target position on the target vehicle is used to attract the target vehicle to drive to the target position.

[0106] In each embodiment of the present application, the potential field of the target vehicle can be determined by calculating the resultant potential field based on any one or more of the repulsive potential field of the road boundary on the target vehicle, the repulsive potential field of the lane line on the target vehicle, the repulsive potential field of the obstacle on the target vehicle, and the gravitational potential field of the target position on the target vehicle. For example, the resultant potential field is calculated based on all the above potential fields, and the potential field of the target vehicle is expressed as follows:

[0107] (6)

[0108] Based on the potential field of the target vehicle, the potential field force of the target vehicle can be obtained by taking partial derivatives in the horizontal and vertical directions, which is expressed as follows:

[0109] (7)

[0110] Based on the potential field force of the target vehicle, the expected yaw angle of the target vehicle in the Cartesian coordinate system, that is, the obstacle avoidance direction, can be obtained, which is expressed as follows:

[0111] (8)

[0112] S13: Based on the obstacle avoidance direction and obstacle information, a navigable danger avoidance corridor is constructed.

[0113] The escape corridor can be understood as a space-time channel for the target vehicle to plan its trajectory. Therefore, the space within the escape corridor should have a low risk of interference with obstacles. For example, based on the obstacle information, the area with a higher risk of collision for the target vehicle can be determined, and the area outside the area with a higher risk can be considered to have a lower risk of collision. Therefore, a suitable range can be delineated in these areas with a lower risk as an escape corridor based on the obstacle avoidance direction.

[0114] In this embodiment, the inaccessible area of ​​the target vehicle can be determined based on the obstacle information, and the first boundary can be determined based on the road boundary or lane line pointed to by the obstacle avoidance direction, and then the target vehicle's avoidance corridor can be constructed based on the first boundary and the inaccessible area.

[0115] The impassable area of ​​the target vehicle can be determined based on a variety of methods. For example, for a static obstacle, the road area occupied by it can be directly used as the impassable area; for a dynamic obstacle, the impassable area of ​​the target vehicle can be determined based on the maximum steering angle and driving speed of the obstacle. In this case, the impassable area of ​​the target vehicle is actually the area that the obstacle can cover within a preset future time period.

[0116] Combined with Figure 3 As shown in the figure, the impassable area of ​​the target vehicle is S2, which is determined by the maximum steering angle △φ of the obstacle. max , the obstacle's driving radius at T time in the future Decision, v agent The determination of the impassable area takes into account the uncertainty of the prediction when the obstacle cuts in urgently. The time T here can be set according to the actual driving environment of the unmanned vehicle, and this application does not limit this.

[0117] The first boundary pointed by the obstacle avoidance direction does not interfere with the impassable area, so the escape corridor can be constructed as an area passing through the first boundary and the impassable area. Figure 3 As shown, for example, the left boundary of the escape corridor may be a preset distance away from the inaccessible area, and the first boundary may be directly used as the right boundary of the escape corridor.

[0118] For the target vehicle, the first boundary pointed by its obstacle avoidance direction can be directly selected as the road boundary; or, it is determined whether the lane line pointed by the obstacle avoidance direction interferes with the impassable area, and the non-interfering lane line with the closest lateral distance to the obstacle is determined as the first boundary; or, the lane lines pointed by the obstacle avoidance direction are screened, and the lane lines with a distance from the impassable area greater than a preset distance are selected, and the lane lines with the closest lateral distance to the obstacle are determined as the first boundary.

[0119] Combined with Figure 4As shown, in this embodiment, during the construction of the danger avoidance corridor, the first reference point can be first determined based on the position of the impassable area closest to the first boundary, and the corresponding second reference point can be determined on the first boundary based on the longitudinal position of the target vehicle invading the obstacle avoidance lane in the obstacle avoidance direction. Subsequently, the first corridor boundary is determined based on the first reference point, and the second corridor boundary is determined based on the second reference point. Finally, based on the first corridor boundary and the second corridor boundary, the danger avoidance corridor of the target vehicle is constructed.

[0120] The obstacle avoidance lane is the lane corresponding to the first boundary, for example, a lane with the first boundary as the left boundary, or a lane with the first boundary as the right boundary, and the present application does not limit this. The first corridor boundary includes a first boundary segment longitudinally adjacent to the target vehicle and a second boundary segment longitudinally away from the target vehicle, the first boundary segment starts from the first reference point and is parallel to the obstacle avoidance direction, and the second boundary segment starts from the first reference point and is parallel to the first boundary. Similarly, the second corridor boundary includes a third boundary segment longitudinally adjacent to the target vehicle and a fourth boundary segment longitudinally away from the target vehicle, the third boundary segment starts from the second reference point and is parallel to the obstacle avoidance direction, and the fourth boundary segment starts from the second reference point and is parallel to the first boundary.

[0121] Continue to combine Figure 4 As shown, the S1 area in the escape corridor can be regarded as the area with the highest driving safety level of the target vehicle, and the S3 area can be regarded as the area with a relatively high driving safety level of the target vehicle. Taking into account the uncertainty of the prediction, in this embodiment, the S2 area and the S4 area are regarded as areas with a lower driving safety level of the target vehicle. It can be seen that the escape corridor in this embodiment is essentially constructed by the S1 area and the S3 area, but this is not a limitation on the area selection of the escape corridor. In some alternative embodiments, the S4 area can also be constructed as part of the escape corridor.

[0122] S14. Planning a driving trajectory of the target vehicle based on the risk avoidance corridor.

[0123] In one embodiment, planning of driving trajectories based on the escape corridor can be regarded as solving the driving trajectory problem with the escape corridor as a constraint. In the solving process, conditions such as the comfort and stability of the vehicle can also be considered at the same time to impose appropriate constraints.

[0124] For example, the problem to be solved can be defined as:

[0125]

[0126] Among them, ψ is the driving trajectory, Γ is the set of plannable driving trajectories, αs is the trajectory safety weight, αt is the trajectory planning time weight, ψ pre is the reference trajectory, the planning time t is the time from the current moment to the target vehicle stopping, ψlow and ψ up are the lower and upper boundaries of the travel trajectory constraints imposed by the hazard avoidance corridor, θ max is the maximum steering angle of the target vehicle, θ i is the heading angle of the ith track point in the driving trajectory, v i is the longitudinal velocity of the ith point of the driving trajectory, u is the road friction coefficient, W is the width of the target vehicle, H is the height of the center of gravity of the target vehicle, R i is the curvature radius of the ith point of the driving trajectory.

[0127] It can be seen that formula (9) is the expected degree of deviation of the driving trajectory from the reference trajectory and the required parking time as small as possible. In formula (13), y(x) is a differentiable planning curve expressed in Cartesian coordinates, that is:

[0128]

[0129] In the specific solution process, the generalized Langrange multiplier method can be introduced, that is, equations (9) to (12) are converted into Langrange functions, which are expressed as follows:

[0130]

[0131] Among them, λ1, λ2, λ3, λ4, and λ5 are the corresponding Lagrange multipliers.

[0132] Based on equation (14), the optimal solution to the above problem can be obtained by solving the equations whose partial derivatives of variables and Lagrange multipliers are 0, which is expressed as follows:

[0133] (15)

[0134] In another embodiment, the driving trajectory of the target vehicle can also be planned using the avoidance corridor based on semantic deduction. Specifically, multiple corresponding driving trajectories can be determined based on multiple preset semantic action sequence templates, and the driving trajectory of the target vehicle can be determined from the multiple driving trajectories with the goal of maximizing the preset reward function.

[0135] The semantic action sequence template includes semantic action directions in multiple planning cycles. In this embodiment, assuming that the target vehicle can choose to turn right, turn left, or maintain the current direction in each planning cycle, the corresponding semantic action direction includes any one of turning right, turning left, and maintaining the action.

[0136] Combination Figure 5As shown, some preset semantic action sequence templates are illustrated, and each semantic action sequence template corresponds to a complete planning cycle of the target vehicle trajectory planning. Exemplarily, 0.1s can be used as a planning cycle, and the same semantic action direction can be corresponded in each 0.5s planning cycle (i.e., action planning cycle). It can be understood that when the planning cycle is short, it may be unnecessary to redefine the semantic direction in each planning cycle, and it will cause too many semantic action sequences to be deduced, which will put a high burden on computing resources.

[0137] Therefore, in this embodiment, when the planning duration of the complete planning cycle is determined, the number of action planning cycles (longer than the trajectory planning cycles of the target vehicle) corresponding to each semantic action sequence template can be determined first, and then the semantic action directions can be arranged and combined in each action planning cycle to obtain all the semantic action sequence templates.

[0138] Combination Figure 6 As shown, in this embodiment, the decision direction corresponding to the obstacle avoidance direction can also be determined first, and when the decision direction is a right turn, multiple preset first-type semantic action sequence templates are determined, and when the decision direction is a left turn, multiple preset second-type semantic action sequence templates are determined. The semantic action directions corresponding to the first-type semantic action sequence templates include right turn, or right turn and action hold, and the semantic action directions corresponding to the second-type semantic action sequence templates include left turn, or left turn and action hold.

[0139] When the target vehicle is avoiding obstacles, especially when braking cannot avoid the risk of collision, it needs to make a decision to turn right or left. After the obstacle avoidance direction is determined, in order to deviate from the current driving direction to the obstacle avoidance direction, the target vehicle will inevitably need to perform a deflection action in the corresponding direction within at least one planning cycle. Corresponding to this embodiment, the semantic action sequence template for semantic deduction of the target vehicle does not include the semantic action direction opposite to the decision direction action, thereby reducing the number of semantic action sequence templates that need to be deduced during trajectory planning.

[0140] Ginseng Figure 7 , the following introduces the specific steps of deducing the trajectory in each planning cycle in this embodiment.

[0141] S141. Determine a front wheel turning angle of the target vehicle in the current planning period based on a preview point of the target vehicle in the danger avoidance corridor in the current planning period.

[0142] The preview point can be determined in a variety of ways. For example, you can first set the forward exploration distance of the preview point in the longitudinal direction, and then determine the corresponding lateral target position. The process is shown as follows:

[0143]

[0144] Among them, v i is the longitudinal velocity of the ith point of the driving trajectory, L tar is the horizontal coordinate of the expected horizontal boundary. t is the set forward exploration time, for example, 1.5s.

[0145] With reference Figure 8 In one embodiment, the desired lateral boundary may be determined by shrinking the obstacle avoidance lane away from the target vehicle by a preset distance. For example, the desired lateral boundary may be determined by shrinking the obstacle avoidance lane away from the target vehicle by 0.6 times the vehicle width.

[0146] Subsequently, the SL coordinates of the preview point can be converted into Cartesian coordinates (x, y), and the Euclidean distance between the current position (x0, y0) of the target vehicle and the preview point can be calculated, which is expressed as follows:

[0147]

[0148] The vector angle from the current state of the target vehicle to the state of the preview point is expressed as follows:

[0149]

[0150] The front wheel steering angle compensation of the target vehicle in the current state is expressed as follows:

[0151]

[0152] Among them, θ0 is the heading angle of the target vehicle in its current state.

[0153] Finally, S' and △α are input into the pure pursuit model (pure pursuit, pp) to calculate the front wheel turning angle of the target vehicle in the current planning cycle, which is expressed as follows:

[0154]

[0155] Where W is the wheelbase of the target vehicle.

[0156] In this embodiment, when determining the front wheel turning angle of the target vehicle in the current planning cycle, kinematic constraints may also be considered. For example, the front wheel turning angle acceleration, and / or the front wheel turning angle jump, and / or the maximum front wheel turning angle are used to constrain the front wheel turning angle of the target vehicle in the current planning cycle. The process is shown as follows:

[0157] Steer at maximum front wheel angle max To constrain:

[0158]

[0159] Calculate the expected jump:

[0160]

[0161] in, To control the lateral acceleration, is the current lateral acceleration of the target vehicle, is the lateral control jump, v' is the longitudinal speed, W is the wheelbase of the target vehicle, K0 is the current curvature of the driving trajectory, v0 is the current speed of the target vehicle, and △t is the execution time of the current action (i.e., the planning cycle length).

[0162] The integral imposes constraints:

[0163]

[0164] in, is the maximum lateral acceleration, is the maximum lateral acceleration.

[0165] The constrained front wheel turning angle is expressed as:

[0166]

[0167] At the maximum front wheel angular velocity ω max To constrain:

[0168]

[0169] Among them, K0 is the current curvature of the driving trajectory, and steer0 is the front wheel steering angle of the target vehicle in the current state.

[0170] Finally, the front wheel turning angle of the target vehicle in the current planning cycle is obtained after constraints:

[0171]

[0172] S142: Based on the obstacle avoidance direction of the target vehicle, determine the lateral obstacle information ahead of the target vehicle.

[0173] S143. Determine the longitudinal speed of the target vehicle in the current planning cycle based on the lateral obstacle information.

[0174] In this embodiment, the lateral obstacle information includes speed information and longitudinal distance information from the target vehicle, and the longitudinal speed of the target vehicle in the current planning cycle can be determined based on the intelligent driver model (IDM). It is expressed as follows:

[0175]

[0176] where a IDMis the longitudinal acceleration of the target vehicle, v is the current speed of the target vehicle, v0 is the expected speed of the target vehicle in free flow state, a is the expected acceleration of the target vehicle, △v is the speed difference between the target vehicle and the obstacle, δ is the acceleration index, which indicates the influence of speed on acceleration, is the desired safe vehicle distance, s is the longitudinal distance between the target vehicle and the obstacle, s0 is the minimum vehicle distance that the target vehicle hopes to maintain in a stationary state, T is the safe headway, b is the deceleration that meets the comfort requirements, v' is the longitudinal speed of the target vehicle, and △t is the execution time of the current action (i.e., the planning cycle length).

[0177] In this embodiment, kinematic constraints may also be considered when determining the longitudinal velocity of the target vehicle in the current planning cycle. For example, the longitudinal velocity of the target vehicle in the current planning cycle is constrained by longitudinal acceleration, and / or longitudinal jump, and / or non-reverse direction. The process is shown as follows:

[0178] The non-reversing direction constraint is used to constrain the target vehicle from reversing during trajectory planning, as shown below:

[0179]

[0180] Calculate the expected jump:

[0181]

[0182] in, is the longitudinal control acceleration, v0 is the current longitudinal velocity of the target vehicle, is the current longitudinal acceleration of the target vehicle.

[0183] The integral imposes constraints:

[0184]

[0185] in, To control the vertical jump, is the maximum longitudinal deceleration speed, is the maximum longitudinal acceleration, is the maximum longitudinal deceleration, is the maximum longitudinal acceleration.

[0186] Therefore, the longitudinal velocity in equation (37) can be expressed as:

[0187]

[0188] S144. Determine the trajectory state information of the target vehicle in the current planning cycle based on the front wheel turning angle and the longitudinal speed of the target vehicle in the current planning cycle.

[0189] Based on the transfer differential equation:

[0190]

[0191] By forward integration, we can obtain:

[0192]

[0193] Among them, S' is the target state of the target vehicle's driving trajectory, K' is the target curvature, is the target longitudinal acceleration.

[0194] Thus, the step of deducing the trajectory state within a planning cycle is completed. Similarly, substitute K' into formula (24) as the current curvature of the driving trajectory, Substituting equation (39) as the current longitudinal acceleration of the target vehicle, the driving trajectory within the complete planning cycle can be obtained by repeated recursion.

[0195] In this embodiment, the preset reward function is constructed based on the danger avoidance corridor. Specifically, the preset reward function may include at least one of a safety distance cost sub-item, a safety speed cost sub-item, and a horizontal and vertical coordination cost sub-item.

[0196] The safety distance cost sub-item represents the offset cost of the driving trajectory relative to the reference trajectory in the safe corridor in each planning cycle. The reference trajectory can be determined in a variety of ways, for example, the center line of the safe corridor can be used as the reference trajectory; or the boundary of the safe corridor away from obstacles can be used as the reference trajectory.

[0197] Combination Figure 8 As shown, in one embodiment, the desired lateral boundary can be determined by shrinking the boundary of the obstacle avoidance lane away from the target vehicle by a preset distance, and the trajectory of the target vehicle pointing to the desired lateral boundary in the obstacle avoidance direction is determined as the first reference trajectory segment, and the segment of the desired lateral boundary away from the reference position in the longitudinal direction is determined as the second reference trajectory segment, and finally based on the first reference trajectory segment and the second reference trajectory segment, the reference trajectory in the danger avoidance corridor is determined. The reference position is the intersection position of the desired lateral boundary and the first reference trajectory segment.

[0198] The safety distance cost sub-item is expressed as:

[0199]

[0200] Among them, α s is the cost weight of the safety distance, and are the positions of the i-th trajectory point in the driving trajectory and the reference trajectory respectively.

[0201] The safety speed cost sub-item represents the longitudinal speed cost of the driving trajectory in each planning cycle, which is expressed as:

[0202]

[0203] Among them, α v is the cost weight of safe speed, v i is the speed of the ith point in the driving trajectory.

[0204] The cost sub-item of the horizontal and vertical coordination represents the difference cost of the longitudinal speed of the target vehicle in each planning cycle relative to the reference longitudinal speed. The reference longitudinal speed is the longitudinal speed corresponding to the smaller of the rollover moment and the sideslip moment of the target vehicle. It can be expressed as:

[0205]

[0206] Among them, gR i u is the rollover moment, gWR i / 2H is the sideslip torque.

[0207] Among the above cost sub-items, the sum of the safety distance cost sub-item and the safety speed cost sub-item can be regarded as the risk cost R risk , the horizontal and vertical coordination cost sub-item can be regarded as the control cost. The preset reward function can, for example, include these costs at the same time, and determine the driving trajectory with the largest reward value corresponding to each solved driving trajectory as the driving trajectory of the target vehicle.

[0208] The trajectory planning method in the above embodiments can be applied to avoid all obstacles in the driving environment. Alternatively, it can be determined whether a collision can be avoided by emergency braking without changing the current driving direction.

[0209] Combination Fig. 9 As shown, in this embodiment, the obstacle risk discrimination model is used to determine whether the collision can be avoided by emergency braking, that is, whether it is necessary to use the above-mentioned trajectory planning method to avoid the obstacle. Specifically, first, the longitudinal speed of the target vehicle and the obstacle, as well as the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle are obtained. And based on the longitudinal speed information of the target vehicle and the braking response delay, the braking distance of the target vehicle is determined; and based on the longitudinal speed information of the obstacle and the risk deceleration, the deceleration distance of the obstacle is determined. Finally, based on the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle, the braking distance of the target vehicle and the deceleration distance of the obstacle, it is determined whether the target vehicle avoids the obstacle.

[0210] The risk discrimination model is expressed as follows:

[0211]

[0212] The logic function sgn(x) is defined as follows:

[0213]

[0214] is the longitudinal velocity of the target vehicle, is the velocity component of the obstacle that cuts in along the direction of the target vehicle’s travel, is the target vehicle response delay, is the maximum braking deceleration of the target vehicle, is the risk deceleration of the obstacle being cut into, It is the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle.

[0215] It can be seen that when sgn=1, it means that the target vehicle still has the risk of collision with the obstacle through emergency braking; conversely, when sgn=0, it means that the target vehicle can avoid the risk of collision with the obstacle through emergency braking.

[0216] Ginseng Fig.10 , an embodiment of the vehicle trajectory planning device of the present application is introduced. In this embodiment, the vehicle trajectory planning device includes an acquisition module 21, a calculation module 22, a construction module 23 and a planning module 24.

[0217] The acquisition module 21 is used to acquire the driving environment information of the target vehicle, wherein the driving environment information includes at least one of road boundary information, lane line information, obstacle information and target position information; the calculation module 22 is used to calculate the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle; the construction module 23 is used to construct a safe corridor passable by the target vehicle based on the obstacle avoidance direction and obstacle information; the planning module 24 is used to plan the driving trajectory of the target vehicle based on the safe corridor.

[0218] In one embodiment, the construction module 23 is specifically used to determine the inaccessible area of ​​the target vehicle based on the obstacle information; determine the first boundary based on the road boundary or lane line pointed to by the obstacle avoidance direction, wherein the first boundary and the inaccessible area do not interfere with each other; and construct a safe haven corridor for the target vehicle based on the first boundary and the inaccessible area.

[0219] In one embodiment, the construction module 23 is specifically used to determine a first reference point based on the position of the impassable area closest to the first boundary; determine a corresponding second reference point on the first boundary based on the longitudinal position of the target vehicle invading the obstacle avoidance lane in the obstacle avoidance direction, wherein the obstacle avoidance lane is the lane corresponding to the first boundary; determine a first corridor boundary based on the first reference point, wherein the first corridor boundary includes a first boundary segment longitudinally adjacent to the target vehicle and a second boundary segment longitudinally away from the target vehicle, the first boundary segment starts from the first reference point and is parallel to the obstacle avoidance direction, and the second boundary segment starts from the first reference point and is parallel to the first boundary; determine a second corridor boundary based on the second reference point, wherein the second corridor boundary includes a third boundary segment longitudinally adjacent to the target vehicle and a fourth boundary segment longitudinally away from the target vehicle, the third boundary segment starts from the second reference point and is parallel to the obstacle avoidance direction, and the fourth boundary segment starts from the second reference point and is parallel to the first boundary; construct a danger avoidance corridor for the target vehicle based on the first corridor boundary and the second corridor boundary.

[0220] In one embodiment, the construction module 23 is specifically used to determine the impassable area of ​​the target vehicle based on the maximum steering angle and driving speed of the obstacle, wherein the impassable area of ​​the target vehicle is the area that the obstacle can cover within a preset future time period.

[0221] In one embodiment, the planning module 24 is specifically used to determine a corresponding plurality of driving trajectories based on a plurality of preset semantic action sequence templates, wherein the semantic action sequence templates include semantic action directions within a plurality of planning cycles, and the semantic action directions include any one of a right turn, a left turn, and action maintenance; and determine a driving trajectory of a target vehicle among the plurality of driving trajectories with the goal of maximizing a preset reward function, wherein the preset reward function is constructed based on a risk avoidance corridor.

[0222] In one embodiment, the planning module 24 is specifically used to determine the decision direction corresponding to the obstacle avoidance direction, wherein the decision direction includes a right turn and a left turn; when the decision direction is a right turn, a plurality of preset first-type semantic action sequence templates are determined, and when the decision direction is a left turn, a plurality of preset second-type semantic action sequence templates are determined, wherein the semantic action directions corresponding to the first-type semantic action sequence templates include a right turn, or a right turn and an action hold, and the semantic action directions corresponding to the second-type semantic action sequence templates include a left turn, or a left turn and an action hold.

[0223] In one embodiment, the planning module 24 is specifically used to determine the front wheel turning angle of the target vehicle in the current planning cycle based on the preview aiming point of the target vehicle in the risk avoidance corridor in the current planning cycle; determine the lateral obstacle information ahead of the target vehicle based on the obstacle avoidance direction of the target vehicle, wherein the lateral obstacle information includes speed information and longitudinal distance information from the target vehicle; determine the longitudinal speed of the target vehicle in the current planning cycle based on the lateral obstacle information; determine the trajectory state information of the target vehicle in the current planning cycle based on the front wheel turning angle and longitudinal speed of the target vehicle in the current planning cycle.

[0224] In one embodiment, the planning module 24 is specifically used to shrink the obstacle avoidance lane away from the boundary of the target vehicle inward by a preset distance to determine the expected lateral boundary, wherein the obstacle avoidance lane is a lane corresponding to a first boundary, the first boundary is a road boundary or lane line pointed to by the obstacle avoidance direction, and the first boundary and the inaccessible area of ​​the target vehicle do not interfere with each other; the trajectory of the target vehicle pointing to the expected lateral boundary in the obstacle avoidance direction is determined as a first reference trajectory segment; the segment of the expected lateral boundary away from the reference position in the longitudinal direction is determined as a second reference trajectory segment, wherein the reference position is the intersection of the expected lateral boundary and the first reference trajectory segment; based on the first reference trajectory segment and the second reference trajectory segment, the reference trajectory in the hazard avoidance corridor is determined.

[0225] In one embodiment, the calculation module 22 is specifically used to calculate the repulsive potential field of the road boundary on the target vehicle based on the road boundary information; calculate the repulsive potential field of the lane line on the target vehicle based on the lane line information; calculate the repulsive potential field of the obstacle on the target vehicle based on the obstacle information; calculate the gravitational potential field of the target position on the target vehicle based on the target position information; calculate the resultant potential field based on the repulsive potential field of the road boundary on the target vehicle, and / or the repulsive potential field of the lane line on the target vehicle, and / or the repulsive potential field of the obstacle on the target vehicle, and / or the gravitational potential field of the target position on the target vehicle, so as to determine the potential field of the target vehicle.

[0226] In one embodiment, a judgment module 25 is further included, which is used to obtain the longitudinal speed of the target vehicle and the obstacle, and the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle; determine the braking distance of the target vehicle based on the longitudinal speed information of the target vehicle and the braking response delay; determine the deceleration distance of the obstacle based on the longitudinal speed information and the risk deceleration of the obstacle; and judge whether the target vehicle avoids the obstacle based on the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle, the braking distance of the target vehicle and the deceleration distance of the obstacle.

[0227] As above Figures 1 to 9 , a vehicle trajectory planning method according to an embodiment of this specification is described. The details mentioned in the above description of the method embodiment are also applicable to the vehicle trajectory planning device of the embodiment of this specification. The above vehicle trajectory planning device can be implemented by hardware, or by software, or a combination of hardware and software.

[0228] Fig.11 FIG. 1 shows a hardware structure diagram of an unmanned vehicle according to an embodiment of the present specification. Fig.11 As shown, the unmanned vehicle 30 may include at least one processor 31, a storage 32 (e.g., a non-volatile memory), a memory 33, and a communication interface 34, and the at least one processor 31, the storage 32, the memory 33, and the communication interface 34 are connected together via an internal bus 35. At least one processor 31 executes at least one computer-readable instruction stored or encoded in the memory 32.

[0229] It should be understood that the computer executable instructions stored in the memory 32, when executed, cause at least one processor 31 to perform the above combined operations in various embodiments of the present specification. Figures 1 to 9 Describes the various operations and functions.

[0230] In the embodiments of the present specification, the unmanned vehicle 30 can be configured with a functional terminal to carry the above-mentioned hardware structure, and the terminal may include but is not limited to: a personal computer, a server computer, a workstation, a desktop computer, a laptop computer, a notebook computer, a mobile electronic device, a smart phone, a tablet computer, a cellular phone, a personal digital assistant (PDA), a handheld device, a messaging device, a wearable electronic device, a consumer electronic device, and the like.

[0231] According to one embodiment, a program product such as a machine-readable medium is provided. The machine-readable medium may have instructions (i.e., the above-mentioned elements implemented in the form of software), which, when executed by a machine, causes the machine to perform the above-mentioned combination of various embodiments of this specification. Figure 1-Figure 9 Specifically, a system or device equipped with a readable storage medium may be provided, on which a software program code implementing the functions of any of the above-mentioned embodiments is stored, and a computer or processor of the system or device reads and executes the instructions stored in the readable storage medium.

[0232] In this case, the program code itself read from the machine-readable medium can realize the function of any one of the above embodiments, and thus the machine-readable code and the machine-readable storage medium storing the machine-readable code constitute part of this specification.

[0233] Examples of readable storage media include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD-RW), magnetic tapes, non-volatile memory cards, and ROMs. Optionally, the program code may be downloaded from a server computer or a cloud via a communication network.

[0234] Those skilled in the art should understand that the various embodiments disclosed above can be modified and altered in various ways without departing from the essence of the invention. Therefore, the protection scope of this specification should be defined by the appended claims.

[0235] It should be noted that not all steps and units in the above-mentioned processes and system structure diagrams are necessary, and some steps or units can be ignored according to actual needs. The execution order of each step is not fixed and can be determined as needed. The device structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some units may be implemented by the same physical client, or some units may be implemented by multiple physical clients, or some components in multiple independent devices may be implemented together.

[0236] In the above embodiments, the hardware unit or module can be implemented mechanically or electrically. For example, a hardware unit, module or processor can include permanent dedicated circuits or logic (such as special processors, FPGA or ASIC) to complete the corresponding operation. The hardware unit or processor can also include programmable logic or circuits (such as general-purpose processors or other programmable processors), which can be temporarily set by software to complete the corresponding operation. The specific implementation method (mechanical method, or dedicated permanent circuit, or temporarily set circuit) can be determined based on cost and time considerations.

[0237] The specific embodiments described above in conjunction with the accompanying drawings describe exemplary embodiments, but do not represent all embodiments that can be implemented or fall within the scope of protection of the claims. The term "exemplary" used throughout this specification means "used as an example, instance or illustration" and does not mean "preferred" or "having advantages" over other embodiments. For the purpose of providing an understanding of the described technology, the specific embodiments include specific details. However, these technologies can be implemented without these specific details. In some instances, in order to avoid making the concepts of the described embodiments difficult to understand, well-known structures and devices are shown in block diagram form.

[0238] The above description of the present disclosure is provided to enable any person of ordinary skill in the art to implement or use the present disclosure. Various modifications to the present disclosure will be apparent to those of ordinary skill in the art, and the general principles corresponding to the present disclosure may be applied to other variations without departing from the scope of protection of the present disclosure. Therefore, the present disclosure is not limited to the examples and designs described herein, but is consistent with the widest range of principles and novel features disclosed herein.

Claims

1. A vehicle trajectory planning method, characterized in that: The method comprises: Acquiring driving environment information of the target vehicle, wherein the driving environment information includes at least one of road boundary information, lane line information, obstacle information, and target position information; Calculating the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle; Based on the obstacle avoidance direction and obstacle information, constructing a danger avoidance corridor passable by the target vehicle; Based on the escape corridor, planning the driving trajectory of the target vehicle; Wherein, based on the obstacle avoidance direction and obstacle information, constructing a danger avoidance corridor accessible to the target vehicle specifically includes: Based on the obstacle information, determining an inaccessible area for the target vehicle; Determine a first boundary based on a road boundary or a lane line pointed to by the obstacle avoidance direction, wherein the first boundary and the impassable area do not interfere with each other; Based on the first boundary and the impassable area, a safe haven corridor for the target vehicle is constructed, where the safe haven corridor passes through an area between the first boundary and the impassable area.

2. The vehicle trajectory planning method according to claim 1, characterized in that: Based on the obstacle avoidance direction and obstacle information, constructing a hazard avoidance corridor accessible to the target vehicle, specifically including: Determining a first reference point based on a position of the impassable area closest to the first boundary; Based on the longitudinal position of the target vehicle intruding into the obstacle avoidance lane in the obstacle avoidance direction, determining a corresponding second reference point on the first boundary, wherein the obstacle avoidance lane is a lane corresponding to the first boundary; Based on the first reference point, determining a first corridor boundary, wherein the first corridor boundary includes a first boundary segment longitudinally adjacent to the target vehicle and a second boundary segment longitudinally away from the target vehicle, the first boundary segment starts from the first reference point and is parallel to the obstacle avoidance direction, and the second boundary segment starts from the first reference point and is parallel to the first boundary; Based on the second reference point, determining a second corridor boundary, wherein the second corridor boundary includes a third boundary segment longitudinally adjacent to the target vehicle and a fourth boundary segment longitudinally away from the target vehicle, the third boundary segment takes the second reference point as a starting point and is parallel to the obstacle avoidance direction, and the fourth boundary segment takes the second reference point as a starting point and is parallel to the first boundary; A danger avoidance corridor for the target vehicle is constructed based on the first corridor boundary and the second corridor boundary.

3. The vehicle trajectory planning method according to claim 1, characterized in that: Determining the impassable area of ​​the target vehicle based on the obstacle information specifically includes: Based on the maximum steering angle and driving speed of the obstacle, the impassable area of ​​the target vehicle is determined, wherein the impassable area of ​​the target vehicle is an area that can be covered by the obstacle within a preset future time period.

4. The vehicle trajectory planning method according to claim 1, characterized in that: Based on the avoidance corridor, planning the driving trajectory of the target vehicle specifically includes: Determine a plurality of corresponding driving trajectories based on a plurality of preset semantic action sequence templates, wherein the semantic action sequence templates include semantic action directions within a plurality of planning cycles, and the semantic action directions include any one of a right turn, a left turn, and action holding; The driving trajectory of the target vehicle is determined among the multiple driving trajectories with the goal of maximizing a preset reward function, wherein the preset reward function is constructed based on a risk avoidance corridor.

5. The vehicle trajectory planning method according to claim 4, characterized in that: Planning the driving trajectory of the target vehicle based on the danger avoidance corridor also includes: Determining a decision direction corresponding to the obstacle avoidance direction, wherein the decision direction includes a right turn and a left turn; When the decision direction is a right turn, multiple preset first-category semantic action sequence templates are determined, and when the decision direction is a left turn, multiple preset second-category semantic action sequence templates are determined, wherein the semantic action directions corresponding to the first-category semantic action sequence templates include right turn, or right turn and action maintenance, and the semantic action directions corresponding to the second-category semantic action sequence templates include left turn, or left turn and action maintenance.

6. The vehicle trajectory planning method according to claim 4, characterized in that: Based on multiple preset semantic action sequence templates, multiple corresponding driving trajectories are determined, including: Determining a front wheel turning angle of the target vehicle in the current planning cycle based on a preview point of the target vehicle in the danger avoidance corridor in the current planning cycle; Based on the obstacle avoidance direction of the target vehicle, determine the lateral obstacle information in front of the target vehicle, wherein the lateral obstacle information includes speed information and longitudinal distance information from the target vehicle; Determining the longitudinal speed of the target vehicle in a current planning cycle based on the lateral obstacle information; Based on the front wheel turning angle and the longitudinal speed of the target vehicle in the current planning period, the trajectory state information of the target vehicle in the current planning period is determined.

7. The vehicle trajectory planning method according to claim 6, characterized in that: The front wheel turning angle of the target vehicle in the current planning cycle is constrained by the front wheel turning angle acceleration, and / or the front wheel turning angle jump, and / or the maximum front wheel turning angle velocity, and / or the maximum front wheel turning angle; and / or, The longitudinal speed of the target vehicle in the current planning cycle is constrained by the longitudinal acceleration, and / or the longitudinal jump, and / or the non-reverse direction.

8. The vehicle trajectory planning method according to claim 4, characterized in that: The preset reward function includes at least one of a safety distance cost sub-item, a safety speed cost sub-item, and a lateral and longitudinal coordination cost sub-item, wherein the safety distance cost sub-item represents the offset cost of the driving trajectory in each planning cycle relative to the reference trajectory in the risk avoidance corridor, the safety speed cost sub-item represents the longitudinal speed cost of the driving trajectory in each planning cycle, and the lateral and longitudinal coordination cost sub-item represents the difference cost of the longitudinal speed of the target vehicle in each planning cycle relative to the reference longitudinal speed, and the reference longitudinal speed is the longitudinal speed corresponding to the smaller of the rollover moment and the sideslip moment of the target vehicle.

9. The vehicle trajectory planning method according to claim 8, characterized in that: The method further comprises: The desired lateral boundary is determined by shrinking the obstacle avoidance lane away from the boundary of the target vehicle by a preset distance, wherein the obstacle avoidance lane is a lane corresponding to a first boundary, the first boundary is a road boundary or lane line pointed by the obstacle avoidance direction, and the first boundary and the impassable area of ​​the target vehicle do not interfere with each other; Determine the trajectory of the target vehicle pointing to the desired lateral boundary in the obstacle avoidance direction as a first reference trajectory segment; Determine a segment of the desired lateral boundary that is longitudinally away from a reference position as a second reference trajectory segment, wherein the reference position is an intersection position of the desired lateral boundary and the first reference trajectory segment; A reference trajectory in the hazard avoidance corridor is determined based on the first reference trajectory segment and the second reference trajectory segment.

10. The vehicle trajectory planning method according to claim 1, characterized in that: Calculating the potential field of the target vehicle based on the driving environment information specifically includes: Based on the road boundary information, calculating the repulsive potential field of the road boundary on the target vehicle; Based on the lane line information, calculating the repulsive potential field of the lane line on the target vehicle; Based on the obstacle information, calculating the repulsive potential field of the obstacle on the target vehicle; Based on the target position information, calculating the gravitational potential field of the target position on the target vehicle; The resultant potential field is calculated based on the repulsive potential field of the road boundary on the target vehicle, and / or the repulsive potential field of the lane line on the target vehicle, and / or the repulsive potential field of the obstacle on the target vehicle, and / or the gravitational potential field of the target position on the target vehicle to determine the potential field of the target vehicle.

11. The vehicle trajectory planning method according to claim 1, characterized in that: The method further comprises: Obtaining the longitudinal speeds of the target vehicle and the obstacle, and the relative distance between the obstacle and the target vehicle in the direction of travel of the target vehicle; Determining a braking distance of the target vehicle based on the longitudinal speed information and the braking response delay of the target vehicle; Determining a deceleration distance of the obstacle based on the longitudinal speed information and the risk deceleration of the obstacle; Based on the relative distance between the obstacle and the target vehicle in the driving direction of the target vehicle, the braking distance of the target vehicle and the deceleration distance of the obstacle, it is determined whether the target vehicle avoids the obstacle.

12. A vehicle trajectory planning device, characterized in that: include: An acquisition module, used to acquire driving environment information of a target vehicle, wherein the driving environment information includes at least one of road boundary information, lane line information, obstacle information and target position information; A calculation module, used to calculate the potential field of the target vehicle based on the driving environment information to determine the obstacle avoidance direction of the target vehicle; A construction module, used to construct a danger avoidance corridor passable by the target vehicle based on the obstacle avoidance direction and obstacle information; A planning module, used for planning the driving trajectory of the target vehicle based on the risk avoidance corridor; Among them, the construction module is specifically used to determine the impassable area of ​​the target vehicle based on the obstacle information; determine the first boundary based on the road boundary or lane line pointed to by the obstacle avoidance direction, wherein the first boundary and the impassable area do not interfere with each other; based on the first boundary and the impassable area, construct a safe corridor for the target vehicle, and the safe corridor is an area passing through the first boundary and the impassable area.

13. An unmanned vehicle, characterized in that: include: at least one processor; as well as A memory storing instructions, which, when executed by the at least one processor, enables the at least one processor to execute the vehicle trajectory planning method according to any one of claims 1 to 11.

14. A machine-readable storage medium storing executable instructions, characterized in that: When the instructions are executed, the machine performs the vehicle trajectory planning method as described in any one of claims 1 to 11.

Images