Vehicle track generation method and device, storage medium and computer equipment

Abstract

The invention discloses a vehicle track generation method and device, a storage medium and computer equipment. The method comprises the following steps: determining a traffic area range of a vehicle and a predicted trajectory of the vehicle according to road information of vehicle driving and obstacle information of obstacles, the predicted trajectory being located in the traffic area range, and the traffic area range and the predicted trajectory being represented by a predetermined coordinate system; the predetermined coordinate system is established based on a road on which the vehicle travels; determining a constraint generation boundary of obstacle avoidance constraint in a predetermined coordinate system according to the predicted trajectory and the traffic area range, the constraint generation boundary being a finite length line located in the traffic area range; generating an obstacle avoidance cost according to the prediction trajectory and the constraint generation boundary; and performing joint space-time optimization on the predicted trajectory according to the obstacle avoidance cost, and generating a target trajectory of the vehicle. The technical problem that a simple and effective trajectory optimization method is lacked when the vehicle is subjected to joint space-time optimization is solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving, and in particular to a vehicle trajectory generation method, device, storage medium and computer equipment. Background Art

[0002] Trajectory planning is a technical link in the autonomous driving system, which converts driving decisions and destinations into trajectories. Trajectory describes the spatial position that the autonomous driving vehicle is expected to reach at each moment in the future, and can be regarded as a curve in space and time. Trajectory planning technology can be roughly divided into space-time decoupled planning and space-time joint planning according to whether the degree of freedom is decomposed. Space-time decoupled planning technology usually decomposes the space-time trajectory into a path and a speed profile, and reduces the difficulty of each of the two solving problems by solving the path and speed separately, but to a certain extent sacrifices the optimality of the final trajectory. Space-time joint planning technology does not perform the above decomposition and directly solves the optimal space-time trajectory, so it can obtain a higher quality trajectory, but it has higher requirements on the adaptability, robustness and computational efficiency of the solver. When obstacles appear in the scene, the process of trajectory solution using space-time joint planning technology in related technologies is more complicated, and requires high computing power, which cannot meet the requirements of real-time and reliability. Moreover, simply avoiding obstacles cannot satisfy users’ requirements for the comfort of the planned trajectory of autonomous vehicles. When the distance between the vehicle and obstacles is too close, users will feel nervous, and such a trajectory obviously cannot satisfy users.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a vehicle trajectory generation method, device, storage medium and computer equipment to at least solve the technical problem of lack of a simple and effective trajectory optimization method when vehicles perform joint spatiotemporal optimization.

[0005] According to one aspect of an embodiment of the present invention, a vehicle trajectory generation method is provided, comprising: determining a passage area range of the vehicle and a predicted trajectory of the vehicle according to road information on which the vehicle is traveling and obstacle information of obstacles, wherein the predicted trajectory is within the passage area range, the passage area range and the predicted trajectory are represented by a predetermined coordinate system, and the predetermined coordinate system is established based on the road on which the vehicle is traveling; determining a constraint generation boundary for avoiding obstacle constraints in the predetermined coordinate system according to the predicted trajectory and the passage area range, wherein the constraint generation boundary is a finite length line located within the passage area range; generating an obstacle avoidance cost according to the predicted trajectory and the constraint generation boundary; and performing joint spatiotemporal optimization on the predicted trajectory according to the obstacle avoidance cost to generate a target trajectory of the vehicle.

[0006] Optionally, generating an obstacle avoidance cost based on the predicted trajectory and the constraint generating boundary includes: determining an ego-vehicle corresponding area of ​​the vehicle at different times based on the predicted trajectory; determining a positional relationship between the ego-vehicle corresponding area and the constraint generating boundary; and determining the obstacle avoidance cost based on the positional relationship.

[0007] Optionally, determining the positional relationship between the area corresponding to the ego vehicle and the constraint generating boundary includes: determining that a portion of the area corresponding to the ego vehicle that is between the constraint generating boundary and the obstacle is an out-of-bounds portion; determining the farthest distance from the constraint generating boundary in the out-of-bounds portion as an out-of-bounds distance corresponding to the area corresponding to the ego vehicle, wherein the positional relationship includes the out-of-bounds distance; determining the obstacle avoidance cost based on the positional relationship includes: determining the obstacle avoidance cost based on the out-of-bounds distance, wherein the obstacle avoidance cost is used to reduce the out-of-bounds distance.

[0008] Optionally, determining the positional relationship between the area corresponding to the ego-vehicle and the constraint generating boundary includes: determining a buffer area between the constraint generating boundary and the vehicle; determining the area of ​​a portion of the area corresponding to the ego-vehicle located within the buffer area, wherein the area is used to characterize the positional relationship; determining the obstacle avoidance cost based on the positional relationship includes: determining the obstacle avoidance cost based on the area, wherein the size of the obstacle avoidance cost is positively correlated with the area.

[0009] Optionally, determining a constraint generating boundary for avoiding obstacle constraints in the predetermined coordinate system according to the predicted trajectory and the range of the passage area includes: when the predetermined coordinate system is a Freynes coordinate system established along a reference line of the road, and the Freynes coordinate system includes an arc major axis and a lateral distance axis, projecting the sampling points on the obstacle onto the reference line of the road according to the obstacle information to obtain a first Freynes coordinate of the sampling points, and projecting the predicted vehicle position of the vehicle at different times onto the reference line of the road according to the predicted trajectory to obtain a second Freynes coordinate of the vehicle; determining the lateral coordinate parameters and length of the constraint generating boundary according to the first Freynes coordinate and the second Freynes coordinate, wherein the lateral coordinate parameters of the constraint generating boundary are expressed by lateral distance axis coordinates, and the length of the constraint generating boundary is expressed by arc major axis coordinates.

[0010] Optionally, determining the length of the constraint generation boundary based on the first Freynes coordinates and the second Freynes coordinates includes: determining the encounter position coordinates of the vehicle and the obstacle when traveling along the predicted trajectory on the arc major axis based on the first Freynes coordinates and the second Freynes coordinates when the vehicle satisfies the advance avoidance condition, wherein the encounter position coordinates are expressed in arc major axis coordinates; determining the advance avoidance position coordinates on the arc major axis based on the encounter position and a preset advance avoidance distance; and adjusting the starting point of the constraint generation boundary from the encounter position coordinates to the advance avoidance position coordinates to obtain the arc length coordinate parameters of the constraint generation boundary.

[0011] Optionally, determining the transverse coordinate parameters of the constraint generation boundary according to the first Freynes coordinate and the second Freynes coordinate includes: determining the transverse coordinate parameters of the initial boundary according to the first Freynes coordinate and the second Freynes coordinate, wherein the transverse coordinate parameters of the initial boundary are expressed by transverse distance axis coordinates; determining whether the type of the obstacle is a dangerous obstacle according to the obstacle information; if the type of the obstacle is a dangerous obstacle, adjusting the transverse coordinate parameters of the initial boundary in a direction away from the obstacle to obtain the transverse coordinate parameters of the constraint generation boundary.

[0012] According to another aspect of an embodiment of the present invention, a vehicle trajectory generating device is also provided, comprising: a first determination module, for determining a passage area range of the vehicle and a predicted trajectory of the vehicle according to road information on which the vehicle is traveling and obstacle information of obstacles, wherein the predicted trajectory is within the passage area range, and the passage area range and the predicted trajectory are represented by a predetermined coordinate system, and the predetermined coordinate system is established based on the road on which the vehicle is traveling; a second determination module, for determining a constraint generation boundary for avoiding obstacle constraints in the predetermined coordinate system according to the predicted trajectory and the passage area range, wherein the constraint generation boundary is a finite length line located within the passage area range; a generation module, for generating an obstacle avoidance cost according to the predicted trajectory and the constraint generation boundary; and an optimization module, for performing joint spatiotemporal optimization on the predicted trajectory according to the obstacle avoidance cost to generate a target trajectory of the vehicle.

[0013] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is further provided, wherein the non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute any one of the above-mentioned vehicle trajectory generation methods.

[0014] According to another aspect of an embodiment of the present invention, a computer device is further provided, comprising a memory and a processor, wherein the memory is used to store programs, and the processor is used to run the programs stored in the memory, wherein when the program is run, any one of the above-mentioned vehicle trajectory generation methods is executed.

[0015] In an embodiment of the present invention, according to the road information of the vehicle and the obstacle information of the obstacle, a predetermined coordinate system established based on the road on which the vehicle is traveling is used to determine the vehicle's travel area range and the vehicle's predicted trajectory, and according to the vehicle's predicted trajectory and the travel area range, a constraint generation boundary for avoiding obstacle constraints is determined, and an obstacle avoidance cost is generated using the predicted trajectory and the constraint generation boundary. The process of joint spatiotemporal optimization is guided by the obstacle avoidance cost, thereby achieving the purpose of efficiently and quickly generating a constraint generation boundary for predictive trajectory optimization, thereby achieving the technical effect of improving the efficiency of solving the vehicle trajectory using the joint spatiotemporal optimization method, and further solving the technical problem of lack of a simple and effective trajectory optimization method when the vehicle performs joint spatiotemporal optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0017] Figure 1A hardware structure block diagram of a computer terminal for implementing a vehicle trajectory generation method is shown;

[0018] Figure 2 is a flow chart of a vehicle trajectory generation method provided according to an embodiment of the present invention;

[0019] Figure 3 is a schematic diagram of using a Fresnel coordinate system to represent obstacles according to an optional embodiment of the present invention;

[0020] Figure 4 is a road schematic diagram represented by a Fresnel coordinate system according to an optional embodiment of the present invention;

[0021] Figure 5 is a schematic diagram of a method for extending a constraint generation boundary according to an optional embodiment of the present invention;

[0022] Figure 6 is a schematic diagram of a double-side avoidance provided according to an optional embodiment of the present invention;

[0023] Figure 7 is a schematic diagram of the principle of generating a boundary generation cost based on constraints according to an optional embodiment of the present invention;

[0024] Figure 8 4 is a structural block diagram of a vehicle trajectory generating device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention 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 invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof 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.

[0027] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following interpretations:

[0028] The Frenet coordinate system, also called the road coordinate system, takes the current position of the vehicle as the origin, and the coordinate axes are perpendicular to each other, divided into the s-axis direction (i.e. the tangent direction along the road reference line, called the arc length direction) and the l-axis direction (i.e. the current normal direction of the reference line, called the lateral direction), and the coordinates are expressed as (s, l). It should be noted that the Frenet coordinate system can change the origin position and the orientation of the coordinate axes in space as the vehicle moves.

[0029] The rectangular coordinate system takes the vehicle's starting position as the origin, and the coordinate axes are perpendicular to each other, divided into the x direction and the y direction, and the coordinates are expressed as (x, y).

[0030] Reinforcement learning (RL), a field in machine learning, emphasizes how to act based on the environment to maximize expected benefits and is widely used in motion control problems. Reinforcement learning is essentially a control algorithm.

[0031] According to an embodiment of the present invention, an embodiment of a vehicle trajectory generation method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0032] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal for implementing a vehicle trajectory generation method. Figure 1 As shown, the computer terminal 10 may include one or more (102a, 102b, ..., 102n are used to illustrate) processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0033] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0034] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the vehicle trajectory generation method in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the vehicle trajectory generation method of the above-mentioned application is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0035] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0036] The vehicle's autonomous driving control system usually includes navigation and positioning functions, object detection, decision planning, V2X communication and vehicle control. Navigation and positioning can obtain the vehicle's accurate position information and semantic map information around the vehicle through high-precision maps and vehicle positioning algorithms. In most applications, the information obtained by the vehicle through sensors such as cameras, lidars, and millimeter-wave radars can detect the status or related information of other objects around it and classify and identify them. The information obtained through navigation and positioning and object detection will be sent to the decision planning module to further calculate the vehicle's trajectory route in the future. The decision-making planning module usually includes task planning, prediction of the trajectory of surrounding objects, behavioral decision-making and trajectory planning. The vehicle can plan the road section information and lane information that need to be passed in the whole process according to the current position, the path points that need to be passed in the future, the end point information and the map information of the road sections; at the same time, the system can also predict the possible future trajectory of the surrounding objects based on the detection results of the surrounding objects and related algorithms; according to the predicted information of the surrounding objects, the position of the vehicle itself and the relevant traffic rules, the behavioral decision module can give a series of behavioral decisions; finally, the trajectory planning will calculate the trajectory that the vehicle needs to execute in the future period of time through numerical optimization methods based on the behavioral decision information given by the behavioral decision.

[0037] Trajectory planning is a technical link in the autonomous driving system, which can convert driving decisions and destinations into trajectories. The trajectory describes the spatial position that the autonomous driving vehicle is expected to reach at each moment in the future, and can be regarded as a curve in space and time. Trajectory planning technology can be roughly divided into space-time decoupling planning and space-time joint planning according to whether the degree of freedom is decomposed. Space-time decoupling planning technology usually decomposes the space-time trajectory into path and speed curves, and reduces the difficulty of each of the two solving problems by solving the path and speed separately, but to a certain extent sacrifices the optimality of the final trajectory. Space-time joint planning technology does not perform the above decomposition and directly solves the optimal space-time trajectory, so it can obtain a higher quality trajectory, but it requires higher adaptability, robustness and computational efficiency of the solver.

[0038] However, when using the spatiotemporal joint planning technology to plan vehicle trajectories, the algorithms for whether to control the vehicle to avoid obstacles and how to control the vehicle to avoid obstacles are relatively complex, resulting in low efficiency in solving the trajectory and often failing to meet the real-time requirements of vehicle trajectory planning.

[0039] To solve the above problems, the present invention proposes a simple and reliable vehicle trajectory generation method to achieve spatiotemporal joint trajectory optimization. Figure 2 FIG. 1 is a flow chart of a vehicle trajectory generation method according to an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0040] Step S202, determining the vehicle's pass area range and the vehicle's predicted trajectory based on the vehicle's road information and obstacle information, wherein the predicted trajectory is within the pass area range, and the pass area range and the predicted trajectory are represented by a predetermined coordinate system, which is established based on the road the vehicle is traveling on.

[0041] In this step, the road information is the information of the road on which the vehicle is currently traveling. The road information may include the width of the road, the number of lanes, the lane width, the lane direction, the spatial position of the road, the position of the lane line, etc. The obstacle may be a static object on the side of the road or on the road, or a passing vehicle on the road. The passing vehicle is an obstacle vehicle for the vehicle. The obstacle information may include the position information of the obstacle and the motion information of the obstacle. The position information may include the current position and posture (abbreviated as posture) of the obstacle, and the motion information may include the current speed, acceleration and other kinematic parameters of the obstacle. Based on the position information and motion information of the obstacle, not only can the obstruction of the obstacle at the current moment be determined, but also the obstacle trajectory of the obstacle in the future period of time can be predicted, which can be applied to the joint spatiotemporal optimization process of the predicted trajectory of the vehicle.

[0042] For example, when the obstacle is an obstacle vehicle, the position information may include the current spatial position of the obstacle vehicle and the current orientation of the obstacle vehicle, and the motion information may include the current speed, acceleration, steering speed and steering acceleration of the obstacle vehicle. Based on the position information and motion information of the obstacle vehicle, the trajectory of the obstacle vehicle in the future can be predicted, so that when predicting the target trajectory of the vehicle, it can be decided whether it is necessary to avoid the obstacle vehicle at a future moment and to what extent the obstacle vehicle should be avoided.

[0043] Optionally, the vehicle's passable area range may be an area range where the vehicle can pass normally in the future without colliding with obstacles. The area range may be determined in the following manner: based on the road information on which the vehicle travels and basic traffic rules, the lane range along which the vehicle can normally travel in the future may be determined; based on the obstacle information, the position of the obstacle at each time in the future may be determined. Then, based on the lane range and the position of the obstacle at each time in the future, the vehicle's passable area range in the future may be generated.

[0044] By establishing a predetermined coordinate system based on the road on which the vehicle is traveling, the effect of quickly determining the constraint generation boundary of the vehicle's avoidance constraint based on a functional relationship can be achieved in the subsequent steps of this embodiment or an optional embodiment, so that the constraint generation boundary is also represented by the predetermined coordinate system. And because the predetermined coordinate system is established based on the road on which the vehicle is traveling, the positional relationship between the constraint generation boundary and the road and between the constraint generation boundary and the trajectory of the vehicle can be quickly determined without the need for complex mathematical operations such as coordinate system substitution, thereby improving the processing speed of the autonomous driving vehicle regarding obstacle avoidance constraints.

[0045] As an optional embodiment, the predetermined coordinate system can be represented by a Frenet coordinate system established along the reference line of the road, and the Frenet coordinate system includes an arc length axis and a lateral distance axis. The origin of the Frenet coordinate system can be located on the reference line of the road, and the reference line of the road can be selected as the central lane line of the road currently traveled by the vehicle. The arc length axis of the Frenet coordinate system is the arc length direction along the reference line, and the lateral distance axis is the distance from the point on the plane projected onto the reference line, wherein the meaning of projecting the point on the plane onto the reference line is to find a projection point on the reference line, connect the projection point on the reference line with the point on the plane, so that the connection line is perpendicular to the tangent of the reference line at the projection point. Using the Frenet coordinate system for calculation in the process of constructing the constraint generation boundary can simplify the complexity of the calculation process, speed up the calculation speed of the constraint generation boundary and the obstacle avoidance cost, and thus enable the vehicle to make a more sensitive avoidance response to obstacles.

[0046] Figure 3 is a schematic diagram of using the Fresnel coordinate system to represent obstacles according to an optional embodiment of the present invention, such as Figure 3 As shown in the figure, the dark rectangle represents the obstacle, the points on the rectangle represent the outline points of the obstacle, and the curve below represents the reference line, which is the arc major axis in the Freynes coordinate system, and the coordinate points are evenly spaced on the arc major axis. For a point on the obstacle outline, it can be projected onto the arc major axis to obtain the arc major axis coordinate of the point (which can be represented by s), and the length of the projected line segment can be measured to obtain the horizontal distance axis coordinate of the point (which can be represented by l), and then the Freynes coordinates of the point on the obstacle outline can be represented as (s, l).

[0047] Step S204, determining a constraint generation boundary of the obstacle avoidance constraint in a predetermined coordinate system according to the predicted trajectory of the vehicle and the range of the pass area, wherein the constraint generation boundary is a finite length line located within the range of the pass area.

[0048] Controlling the vehicle to move forward along the initial predicted trajectory of the vehicle within the passage area cannot meet the needs of users. For example, when the vehicle is too close to an obstacle, the passengers on the vehicle will feel nervous. Therefore, the solution of the present invention can control the vehicle to maintain a certain distance from the obstacle so that the passengers can get a better riding experience. The constraint condition used to guide the vehicle to maintain a certain safe distance from the obstacle in this application is called the obstacle avoidance constraint. It should be noted that in order to guide the vehicle to maintain a distance from the obstacle during the joint spatiotemporal optimization, a constraint generation boundary can be constructed within the passage area to guide the vehicle to reasonably stay away from the obstacle. It should be noted that the constraint generation boundary can indicate that when the position of the vehicle crosses the constraint generation boundary or when the position of the vehicle is close to the constraint generation boundary, an obstacle avoidance cost that constrains the distance between the vehicle and the obstacle in the joint spatiotemporal optimization can be generated, and when the vehicle does not cross the constraint generation boundary or the vehicle is far from the constraint generation boundary, there is no need to generate an obstacle avoidance cost that constrains the distance between the vehicle and the obstacle. Since the boundary is used to determine whether to generate related constraints, the boundary is called the constraint generation boundary.

[0049] Among them, the constraint generation boundary is used to generate an avoidance constraint when executing the joint spatiotemporal optimization algorithm of the vehicle trajectory. The avoidance constraint can be expressed as an obstacle avoidance cost. The optimized vehicle trajectory does not strictly prohibit the vehicle from crossing the constraint generation boundary. Instead, when the vehicle crosses the constraint generation boundary in the initial predicted trajectory, an obstacle avoidance cost will be generated. The obstacle avoidance cost can guide the joint spatiotemporal optimization process so that the vehicle returns to the boundary as much as possible when crossing the constraint generation boundary. As for whether the optimized target trajectory of the vehicle completely returns to the constraint generation boundary, it also depends on other constraint costs in the joint spatiotemporal optimization process.

[0050] It should also be noted that the lateral coordinate parameters and arc length coordinate parameters of the constraint generating boundary can be combined to represent the absolute position of the constraint generating boundary in space. The lateral coordinate parameters of the constraint generating boundary can characterize the distance between each point on the constraint generating boundary and the arc length axis of the Fresnel coordinate system, and the arc length coordinate parameters of the constraint generating boundary can characterize the arc length axis coordinates of each point on the boundary when projected onto the arc length axis. Optionally, when there are multiple obstacles that need to be avoided in the scene, multiple constraint generating boundaries can be determined, and the multiple constraint generating boundaries are not directly connected, and the length of each constraint generating boundary is limited, for example, it is only located between its corresponding obstacle and the lane reference line. The lateral coordinate parameters of the constraint generating boundary represent the lateral offset of the boundary from the lane, and the arc length coordinate parameters can represent the length of the constraint generating boundary extending along the lane direction, wherein the lane direction represents the positive direction of the lane, and the lateral offset of the lane can be the perpendicular direction of the tangent direction at any point in the lane in the lane plane.

[0051] Optionally, the constraint generating boundary can be a line segment, a curve, or a combination of a line segment and a curve, so it can be called a finite length line. Since the predicted trajectory of the vehicle basically extends along the lane line, and obstacles are distributed on both sides of the lane, the Freynes coordinate system can more simply characterize the predicted trajectory of the vehicle, the range of the vehicle's passage area, and the constraint generating boundary. Using the Cartesian coordinate system to characterize the above concepts requires a more cumbersome method to solve the reasonable position of the constraint generating boundary. Using the Freynes coordinate system to characterize the above concepts can naturally generate the Freynes coordinates of the constraint generating boundary based on the preset constraint generating boundary generation logic, thereby speeding up the generation speed of the obstacle avoidance cost and improving the efficiency of vehicle trajectory optimization.

[0052] As an optional embodiment, when the predetermined coordinate system is a Freynes coordinate system established along a reference line of the road, and the Freynes coordinate system includes an arc-length axis and a lateral distance axis, the lateral coordinate parameters and arc-length coordinate parameters of the constraint generation boundary within the passage area can be determined according to the predicted trajectory of the vehicle and the obstacle information in the following manner: the sampling points on the obstacle are projected onto the reference line of the road according to the obstacle information to obtain the first Freynes coordinates of the sampling points, and the predicted position of the vehicle at different times is projected onto the reference line of the road according to the predicted trajectory to obtain the second Freynes coordinates of the vehicle; the lateral coordinate parameters and arc-length coordinate parameters of the constraint generation boundary are determined according to the first Freynes coordinates and the second Freynes coordinates, wherein the lateral coordinate parameters of the constraint generation boundary are expressed by the lateral distance axis coordinates, and the arc-length coordinate parameters of the constraint generation boundary are expressed by the arc-length axis coordinates.

[0053] In this optional embodiment, the sampling points of the obstacle may be points on the outline of the obstacle, and the outline of the obstacle is the outline facing the side of the road where the vehicle is located. Using the Freynes coordinate system to determine the constraint generation boundary can greatly simplify the representation of the constraint generation boundary, making the obstacle avoidance cost generated according to the constraint generation boundary more concise, thereby enabling the target trajectory to be generated more quickly when performing joint spatiotemporal optimization.

[0054] Figure 4 is a road schematic diagram represented by a Fresnel coordinate system according to an optional embodiment of the present invention, such as Figure 4 As shown, based on the road boundary (i.e., the lane boundary in the figure), a fishbone diagram with the road centerline as the reference line can be generated. The points on the centerline can be understood as the scales on the arc major axis of the Freynes coordinate system. The obstacle is the rectangular box in the figure. The boundary of the obstacle can be determined by the sampling points on the contour of the obstacle facing the side of the road where the vehicle is located. By projecting the sampling points onto the reference line, the Freynes coordinates (s, l) of the sampling points can be obtained.

[0055] As an optional embodiment, the arc length coordinate parameters of the constraint generation boundary are determined according to the first Freynes coordinates and the second Freynes coordinates. The following method can be used: when the vehicle and the obstacle meet the advance avoidance conditions, the encounter position coordinates of the vehicle and the obstacle when traveling along the predicted trajectory are determined on the arc long axis according to the first Freynes coordinates and the second Freynes coordinates, wherein the encounter position coordinates are expressed in arc long axis coordinates; the advance avoidance position coordinates are determined on the arc long axis according to the encounter position coordinates and a preset advance avoidance distance; and the starting point of the constraint generation boundary is extended from the encounter position coordinates to the advance avoidance position coordinates to obtain the arc length coordinate parameters of the constraint generation boundary.

[0056] In this optional embodiment, the early avoidance condition may include at least one of the following types of conditions: obstacle volume condition, obstacle speed condition, and obstacle distribution condition. It is understandable that when the obstacle is large in volume, fast in speed, or has different distribution patterns and densities, the psychological feelings of the passengers when the vehicle and the obstacle meet are different. Therefore, this optional embodiment can provide different avoidance strategies for different scenarios. For example, if the obstacle is an obstacle vehicle, and the obstacle vehicle is a relatively dangerous large vehicle, the people on the vehicle are more likely to feel insecure and hope to stay away from the large vehicle; or if the speed difference between the obstacle vehicle and the vehicle is too large, the people on the vehicle will also hope that the vehicle and the meeting vehicle will keep a distance to improve the sense of security. Optionally, vehicles with a length greater than 6 meters, a height greater than 2.2 meters, and a width greater than 2.2 meters can be defined as dangerous large vehicles; vehicles with a speed difference greater than 15km / h with the vehicle can be defined as dangerous vehicles. Therefore, when the scene meets the early avoidance conditions, the length of the constraint generation boundary can be extended through this optional embodiment, so that the target trajectory of the vehicle can start to avoid the dangerous obstacle in advance before it intersects with the dangerous obstacle, and can also delay the end of avoidance of the obstacle after the intersection of the dangerous obstacle is completed. The end point coordinates of the constraint generation boundary are adjusted backward a predetermined distance from the coordinates of the position where the vehicle and the obstacle leave each other, and the adjusted end point position of the constraint generation boundary is the delayed avoidance position coordinate, thereby improving the sense of safety of the driver and passengers.

[0057] Figure 5 is a schematic diagram of a method for extending a constraint generation boundary according to an optional embodiment of the present invention, such as Figure 5 As shown, line segment AB represents Figure 5 The constraint generation boundary corresponding to the obstacle in , AB is extended to a certain length in both directions, namely the "advance avoidance" part and the "delayed end" part. The vehicle in the figure is traveling from left to right, and the encounter position between the vehicle and the obstacle is the right end point of the "advance avoidance" position. The length of the "advance avoidance" part matches the preset avoidance distance, and the avoidance direction is the direction of the arrow in the figure. The dotted line represents the outer lane line of the lane where the vehicle is traveling.

[0058] Optionally, when obstacles appear on both sides of the road of the vehicle, constraint generation boundaries can be generated on both sides of the vehicle respectively, so that the vehicle can travel in the center without deviation. Figure 6 is a schematic diagram of a double-side avoidance according to an optional embodiment of the present invention, such as Figure 6 As shown in the figure, the vehicle meets the upper obstacle vehicle at time T1, and a constraint generation boundary is generated between the vehicle and the upper obstacle vehicle, causing the vehicle's trajectory to deviate downward to avoid the upper obstacle vehicle in the figure; at time T2, the vehicle's trajectory has been avoided to a position far enough to correct the vehicle's front direction; at time T3, the vehicle meets the lower obstacle vehicle, and at this time, constraint generation boundaries are generated between the vehicle and both the upper and lower obstacle vehicles, causing the vehicle to center and return to the center, passing between the two vehicles. When the vehicle surpasses the upper obstacle vehicle, the vehicle's trajectory also exceeds the constraint generation boundary between it and the upper obstacle vehicle. At this time, the vehicle only has a constraint generation boundary with the lower obstacle vehicle, causing the constraint generation boundary to push the vehicle to deviate laterally upward to avoid the lower obstacle vehicle.

[0059] As an optional embodiment, the transverse coordinate parameters of the constraint generation boundary are determined according to the first Freynes coordinate and the second Freynes coordinate, including the following methods: determining the initial boundary transverse coordinate parameters according to the first Freynes coordinate and the second Freynes coordinate, wherein the initial boundary transverse coordinate parameters are represented by transverse distance axis coordinates; determining whether the type of obstacle is a dangerous obstacle according to the obstacle information; if the type of obstacle is a dangerous obstacle, adjusting the initial boundary transverse coordinate parameters in a direction away from the obstacle to obtain the transverse coordinate parameters of the constraint generation boundary.

[0060] According to this optional embodiment, the lateral coordinate parameters of the constraint generation boundary can be determined according to the degree of danger of the obstacle. For example, when the obstacle is more dangerous, the lateral coordinate parameters of the constraint generation boundary are pushed toward the center line of the lane, so that the generated target trajectory can be further away from the obstacle, bringing a better obstacle avoidance experience to the people in the car. Among them, whether the type of obstacle is a dangerous obstacle can be determined based on the obstacle volume condition and / or the obstacle speed condition. When the obstacle volume is large, for example, the obstacle vehicle is a dangerous large vehicle, or the speed difference between the obstacle and the vehicle is too large, for example, the obstacle vehicle is a high-speed oncoming vehicle, then the obstacle is judged to be a dangerous obstacle, and the initial constraint generation boundary is moved in the direction away from the dangerous obstacle.

[0061] Step S206, generating an obstacle avoidance cost according to the predicted trajectory and the constraint generation boundary. The obstacle avoidance cost can be used to guide the calculation results of the algorithm in the joint spatiotemporal optimization, so that the optimized target trajectory can avoid crossing the constraint generation boundary as much as possible.

[0062] As an optional embodiment, the following steps can be used to generate an obstacle avoidance cost based on the predicted trajectory and the constraint generation boundary: determine the corresponding area of ​​the vehicle at different times based on the predicted trajectory; determine the positional relationship between the corresponding area of ​​the vehicle and the constraint generation boundary; and determine the obstacle avoidance cost based on the positional relationship.

[0063] Optionally, the area corresponding to the vehicle may depend on the way the vehicle is modeled, such as modeling the vehicle as a rectangular area or modeling the vehicle as two parallel circular areas. Figure 7 is a schematic diagram of the principle of generating a boundary generation cost based on constraints according to an optional embodiment of the present invention, Figure 7 In the example, the obstacle is modeled as a rectangular box on the left, and the ego vehicle corresponding area of ​​the ego vehicle is modeled as a parallel circle on the right. The positional relationship between the ego vehicle corresponding area and the constraint generation boundary can be a geometric relationship between the two, such as whether the constraint generation boundary passes through the ego vehicle corresponding area, the degree of passing, or the distance between the constraint generation boundary and the ego vehicle corresponding area.

[0064] As an optional embodiment, the positional relationship between the ego-vehicle corresponding area and the constraint generating boundary can be determined in the following manner: determine the portion of the ego-vehicle corresponding area located between the constraint generating boundary and the obstacle as the out-of-bounds portion; determine the farthest distance from the constraint generating boundary in the out-of-bounds portion as the out-of-bounds distance corresponding to the ego-vehicle corresponding area, wherein the positional relationship includes the out-of-bounds distance; determine the obstacle avoidance cost based on the positional relationship, including: determine the obstacle avoidance cost based on the out-of-bounds distance, wherein the obstacle avoidance cost is used to reduce the out-of-bounds distance.

[0065] As an optional embodiment, the positional relationship between the area corresponding to the ego vehicle and the constraint generation boundary can also be determined in the following manner: determine a buffer area between the constraint generation boundary and the vehicle; determine the area of ​​the portion of the area corresponding to the ego vehicle that is located in the buffer area, wherein the area is used to characterize the positional relationship; determine the obstacle avoidance cost based on the positional relationship, including: determine the obstacle avoidance cost based on the area, wherein the size of the obstacle avoidance cost is positively correlated with the area.

[0066] The above two optional embodiments provide two different ways to generate obstacle avoidance costs. Through the above optional embodiments, the state of the vehicle's predicted trajectory and the constraint generation boundary can be quickly determined. The effect of the obstacle avoidance cost is like a "thrust". When the vehicle deviates in the direction of the constraint generation boundary, the obstacle avoidance cost can "push" the vehicle away from the constraint generation boundary in the joint spatiotemporal optimization, so that the optimized target trajectory can be far away from the obstacle.

[0067] Step S208: performing joint spatiotemporal optimization on the predicted trajectory according to the obstacle avoidance cost to generate a target trajectory of the vehicle.

[0068] In this step, space-time joint trajectory optimization is a technology of space-time joint planning. This technology solves the optimal space-time trajectory in the form of numerical optimization. Numerical optimization relies on a cost function to describe the optimization goal. The cost function is a function defined in the trajectory space, which is used to describe the quality of the trajectory. The cost function must be able to fully analyze and evaluate all aspects of the trajectory, including feasibility, safety, comfort, and anthropomorphism. Among them, avoiding obstacles comfortably, smoothly and safely is the basic ability of the trajectory. At the same time, since numerical optimization calculates the optimal trajectory through information such as the value and gradient of the cost function, the cost function needs to be as efficient and concise as possible. In the context of space-time joint trajectory optimization, the present invention proposes a method for constructing a large vehicle avoidance cost function, which can make the trajectory safe, comfortable, and executable for large vehicle avoidance actions at night.

[0069] Optionally, the joint spatiotemporal optimization can be implemented by reinforcement learning (RL), which is a field in machine learning that emphasizes how to act based on the environment to maximize expected benefits. It is widely used in motion control problems and is essentially a control algorithm. Using the reinforcement learning algorithm, the trajectory can be optimized by adjusting the initial reference trajectory of the target vehicle according to the trajectory cost of the target vehicle (such as the obstacle avoidance cost), and the optimized trajectory is used as the new reference trajectory for multiple iterations until the number of iterations reaches a threshold or the trajectory cost converges, thereby obtaining the final optimized trajectory, which is the target trajectory of the vehicle.

[0070] Through the above steps, the vehicle's travel area and the vehicle's predicted trajectory are determined by using a predetermined coordinate system established based on the road on which the vehicle is traveling according to the road information and the obstacle information of the obstacle, and the constraint generation boundary of the obstacle avoidance constraint is determined according to the predicted trajectory and the travel area of ​​the vehicle, and the obstacle avoidance cost is generated by using the predicted trajectory and the constraint generation boundary. The process of joint spatiotemporal optimization is guided by the obstacle avoidance cost, and the purpose of efficiently and quickly generating the constraint generation boundary for predictive trajectory optimization is achieved, thereby achieving the technical effect of improving the efficiency of solving the vehicle trajectory by using the joint spatiotemporal optimization method, and further solving the technical problem of lack of a simple and effective trajectory optimization method when the vehicle performs joint spatiotemporal optimization.

[0071] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0072] Through the description of the above implementation methods, those skilled in the art can clearly understand that the vehicle trajectory generation method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), including a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0073] According to an embodiment of the present invention, a vehicle trajectory generating device for implementing the above vehicle trajectory generating method is also provided. Figure 8 is a structural block diagram of a vehicle trajectory generating device provided according to an embodiment of the present invention. Figure 8 As shown, the vehicle trajectory generating device includes: a first determining module 82, a second determining module 84, a generating module 86 and an optimizing module 88. The vehicle trajectory generating device is described below.

[0074] A first determination module 82 is used to determine the range of the vehicle's pass area and the predicted trajectory of the vehicle according to the road information of the vehicle and the obstacle information of the obstacle, wherein the predicted trajectory is within the pass area, and the pass area and the predicted trajectory are represented by a predetermined coordinate system, which is established based on the road the vehicle is traveling on;

[0075] A second determination module 84, connected to the first determination module 82, is used to determine a constraint generation boundary of the obstacle avoidance constraint according to the predicted trajectory and the pass area range in a predetermined coordinate system, wherein the constraint generation boundary is a finite length line located within the pass area range;

[0076] A generating module 86, connected to the second determining module 84, for generating a boundary and an obstacle avoidance cost according to the predicted trajectory and the constraints;

[0077] The optimization module 88 is connected to the generation module 86 and is used to perform joint spatiotemporal optimization on the predicted trajectory according to the obstacle avoidance cost to generate a target trajectory of the vehicle.

[0078] It should be noted that the first determination module 82, the second determination module 84, the generation module 86 and the optimization module 88 correspond to steps S202 to S208 in the embodiment, and the examples and application scenarios implemented by the multiple modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the above modules, as part of the device, can be run in the computer terminal 10 provided in the embodiment.

[0079] An embodiment of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one network device among multiple network devices of a computer network. The computer device includes a memory and a processor.

[0080] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the vehicle trajectory generation method and device in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, the above-mentioned vehicle trajectory generation method is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0081] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: determine the vehicle's pass area range and the vehicle's predicted trajectory based on the road information of the vehicle and the obstacle information of the obstacle, wherein the predicted trajectory is within the pass area range, the pass area range and the predicted trajectory are represented by a predetermined coordinate system, and the predetermined coordinate system is established based on the road the vehicle is traveling on; determine the constraint generation boundary of the obstacle avoidance constraint in the predetermined coordinate system based on the predicted trajectory and the pass area range, wherein the constraint generation boundary is a finite length line located within the pass area range; generate an obstacle avoidance cost based on the predicted trajectory and the constraint generation boundary; and perform joint spatiotemporal optimization on the predicted trajectory based on the obstacle avoidance cost to generate the vehicle's target trajectory.

[0082] Optionally, the processor may also execute program code for the following steps: generating a boundary based on a predicted trajectory and constraints, and generating an obstacle avoidance cost, including: determining the corresponding area of ​​the vehicle at different times based on the predicted trajectory; determining the positional relationship between the corresponding area of ​​the vehicle and the boundary generated by the constraints; and determining the obstacle avoidance cost based on the positional relationship.

[0083] Optionally, the processor may also execute program code for the following steps: determining the positional relationship between the area corresponding to the ego vehicle and the constraint generating boundary, including: determining the portion of the area corresponding to the ego vehicle that is between the constraint generating boundary and the obstacle as the out-of-bounds portion; determining the farthest distance from the constraint generating boundary in the out-of-bounds portion as the out-of-bounds distance corresponding to the area corresponding to the ego vehicle, wherein the positional relationship includes the out-of-bounds distance; determining the obstacle avoidance cost based on the positional relationship, including: determining the obstacle avoidance cost based on the out-of-bounds distance, wherein the obstacle avoidance cost is used to reduce the out-of-bounds distance.

[0084] Optionally, the processor may also execute program code for the following steps: determining the positional relationship between the area corresponding to the ego vehicle and the constraint generating boundary, including: determining a buffer area between the constraint generating boundary and the vehicle; determining the area of ​​the portion of the area corresponding to the ego vehicle that is located within the buffer area, wherein the area is used to characterize the positional relationship; determining an obstacle avoidance cost based on the positional relationship, including: determining the obstacle avoidance cost based on the area, wherein the size of the obstacle avoidance cost is positively correlated with the area.

[0085] Optionally, the processor may also execute the program code of the following steps: determining a constraint generation boundary of an obstacle avoidance constraint in a predetermined coordinate system according to a predicted trajectory and a range of a passage area, including: when the predetermined coordinate system is a Freynes coordinate system established along a reference line of a road, and the Freynes coordinate system includes an arc major axis and a lateral distance axis, projecting sampling points on the obstacle onto the reference line of the road according to obstacle information to obtain a first Freynes coordinate of the sampling point, and projecting the predicted position of the vehicle at different times onto the reference line of the road according to the predicted trajectory to obtain a second Freynes coordinate of the vehicle; determining the lateral coordinate parameters and length of the constraint generation boundary according to the first Freynes coordinate and the second Freynes coordinate, wherein the lateral coordinate parameters of the constraint generation boundary are expressed by lateral distance axis coordinates, and the length of the constraint generation boundary is expressed by arc major axis coordinates.

[0086] Optionally, the processor may also execute the program code of the following steps: determining the length of the constraint generation boundary according to the first Freynes coordinate and the second Freynes coordinate, including: when the vehicle and the obstacle meet the advance avoidance condition, determining the encounter position coordinates of the vehicle with the obstacle when traveling along the predicted trajectory on the arc major axis according to the first Freynes coordinate and the second Freynes coordinate, wherein the encounter position coordinates are expressed in arc major axis coordinates; determining the advance avoidance position coordinates on the arc major axis according to the encounter position and a preset advance avoidance distance; and adjusting the starting point of the constraint generation boundary from the encounter position coordinates to the advance avoidance position coordinates to obtain the arc length coordinate parameters of the constraint generation boundary.

[0087] Optionally, the processor may also execute the following steps of program code: determining the transverse coordinate parameters of the constraint generating boundary based on the first Freynes coordinate and the second Freynes coordinate, including: determining the transverse coordinate parameters of the initial boundary based on the first Freynes coordinate and the second Freynes coordinate, wherein the transverse coordinate parameters of the initial boundary are expressed by transverse distance axis coordinates; determining whether the type of obstacle is a dangerous obstacle based on obstacle information; if the type of obstacle is a dangerous obstacle, adjusting the transverse coordinate parameters of the initial boundary in a direction away from the obstacle to obtain the transverse coordinate parameters of the constraint generating boundary.

[0088] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a non-volatile storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0089] The embodiment of the present invention further provides a non-volatile storage medium. Optionally, in this embodiment, the non-volatile storage medium can be used to store the program code executed by the vehicle trajectory generation method provided in the above embodiment.

[0090] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0091] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: determining the vehicle's pass area range and the vehicle's predicted trajectory based on the vehicle's travel road information and obstacle information, wherein the predicted trajectory is within the pass area range, the pass area range and the predicted trajectory are represented by a predetermined coordinate system, and the predetermined coordinate system is established based on the road the vehicle is traveling on; determining a constraint generation boundary for avoiding obstacle constraints in the predetermined coordinate system based on the predicted trajectory and the pass area range, wherein the constraint generation boundary is a finite length line located within the pass area range; generating an obstacle avoidance cost based on the predicted trajectory and the constraint generation boundary; and performing joint spatiotemporal optimization on the predicted trajectory based on the obstacle avoidance cost to generate a target trajectory for the vehicle.

[0092] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: generating a boundary based on a predicted trajectory and constraints, and generating an obstacle avoidance cost, including: determining the corresponding area of ​​the vehicle at different times based on the predicted trajectory; determining the positional relationship between the corresponding area of ​​the vehicle and the boundary generated by the constraints; and determining the obstacle avoidance cost based on the positional relationship.

[0093] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: determining the positional relationship between the area corresponding to the ego vehicle and the constraint generating boundary, including: determining the portion of the area corresponding to the ego vehicle that is between the constraint generating boundary and the obstacle as the out-of-bounds portion; determining the farthest distance from the constraint generating boundary in the out-of-bounds portion as the out-of-bounds distance corresponding to the area corresponding to the ego vehicle, wherein the positional relationship includes the out-of-bounds distance; determining the obstacle avoidance cost based on the positional relationship, including: determining the obstacle avoidance cost based on the out-of-bounds distance, wherein the obstacle avoidance cost is used to reduce the out-of-bounds distance.

[0094] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: determining the positional relationship between the area corresponding to the ego vehicle and the constraint generating boundary, including: determining a buffer area between the constraint generating boundary and the vehicle; determining the area of ​​the portion of the area corresponding to the ego vehicle that is located within the buffer area, wherein the area is used to characterize the positional relationship; determining an obstacle avoidance cost based on the positional relationship, including: determining the obstacle avoidance cost based on the area, wherein the size of the obstacle avoidance cost is positively correlated with the area.

[0095] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: determining a constraint generation boundary for avoiding obstacle constraints based on a predicted trajectory and a range of a passable area, including: determining a constraint generation boundary for avoiding obstacle constraints in a predetermined coordinate system based on the predicted trajectory and the range of a passable area, including: when the predetermined coordinate system is a Freynes coordinate system established along a reference line of a road, and the Freynes coordinate system includes an arc major axis and a lateral distance axis, projecting sampling points on the obstacle onto the reference line of the road according to obstacle information to obtain first Freynes coordinates of the sampling points, and projecting the predicted position of the vehicle at different times of the vehicle onto the reference line of the road according to the predicted trajectory to obtain second Freynes coordinates of the vehicle; determining lateral coordinate parameters and length of the constraint generation boundary based on the first Freynes coordinates and the second Freynes coordinates, wherein the lateral coordinate parameters of the constraint generation boundary are expressed by lateral distance axis coordinates, and the length of the constraint generation boundary is expressed by arc major axis coordinates.

[0096] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: determining the length of the constraint generation boundary based on the first Freynes coordinate and the second Freynes coordinate, including: when the vehicle and the obstacle meet the advance avoidance condition, determining the encounter position coordinates of the vehicle with the obstacle when traveling along the predicted trajectory on the arc major axis based on the first Freynes coordinate and the second Freynes coordinate, wherein the encounter position coordinates are expressed in arc major axis coordinates; determining the advance avoidance position coordinates on the arc major axis based on the encounter position and a preset advance avoidance distance; adjusting the starting point of the constraint generation boundary from the encounter position coordinates to the advance avoidance position coordinates to obtain the arc length coordinate parameters of the constraint generation boundary.

[0097] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: determining the transverse coordinate parameters of the constraint generation boundary based on the first Freynes coordinate and the second Freynes coordinate, including: determining the transverse coordinate parameters of the initial boundary based on the first Freynes coordinate and the second Freynes coordinate, wherein the transverse coordinate parameters of the initial boundary are expressed by transverse distance axis coordinates; determining whether the type of obstacle is a dangerous obstacle based on obstacle information; in the case that the type of obstacle is a dangerous obstacle, adjusting the transverse coordinate parameters of the initial boundary in a direction away from the obstacle to obtain the transverse coordinate parameters of the constraint generation boundary.

[0098] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0099] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0101] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0102] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0103] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0104] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A vehicle trajectory generation method, characterized in that: include: Determine the range of the vehicle's passing area and the predicted trajectory of the vehicle according to the road information of the vehicle and the obstacle information of the obstacle, wherein the predicted trajectory is within the range of the passing area, and the passing area and the predicted trajectory are represented by a predetermined coordinate system, and the predetermined coordinate system is established based on the road the vehicle is traveling on; Determining a constraint generation boundary of an obstacle avoidance constraint in the predetermined coordinate system according to the predicted trajectory and the range of the passage area, wherein the constraint generation boundary is a finite length line located within the range of the passage area; generating a boundary according to the predicted trajectory and the constraint, and generating an obstacle avoidance cost; The predicted trajectory is jointly optimized in time and space according to the obstacle avoidance cost to generate a target trajectory of the vehicle.

2. The method according to claim 1, characterized in that: The generating a boundary according to the predicted trajectory and the constraint and generating an obstacle avoidance cost includes: Determining the corresponding area of ​​the vehicle at different times according to the predicted trajectory; Determining a positional relationship between the ego-vehicle corresponding area and the constraint generation boundary; The obstacle avoidance cost is determined according to the positional relationship.

3. The method according to claim 2, characterized in that The determining of the positional relationship between the ego-vehicle corresponding area and the constraint generation boundary includes: Determine a portion of the ego-vehicle corresponding area between the constraint generation boundary and the obstacle as an out-of-bounds portion; Determine the farthest distance from the constraint generation boundary in the out-of-bounds portion as the out-of-bounds distance corresponding to the area corresponding to the ego-vehicle, wherein the positional relationship includes the out-of-bounds distance; Determining the obstacle avoidance cost according to the position relationship includes: The obstacle avoidance cost is determined according to the cross-border distance, wherein the obstacle avoidance cost is used to reduce the cross-border distance.

4. The method according to claim 2, characterized in that: The determining of the positional relationship between the ego-vehicle corresponding area and the constraint generation boundary includes: determining a buffer region between the constraint generating boundary and the vehicle; Determine the area of ​​the portion of the ego-vehicle corresponding area that is located in the buffer area, wherein the area is used to characterize the positional relationship; Determining the obstacle avoidance cost according to the position relationship includes: The obstacle avoidance cost is determined according to the area of ​​the region, wherein the size of the obstacle avoidance cost is positively correlated with the area of ​​the region.

5. The method according to any one of claims 1 to 4, characterized in that: The step of determining a constraint generation boundary of an obstacle avoidance constraint in the predetermined coordinate system according to the predicted trajectory and the range of the passage area includes: In a case where the predetermined coordinate system is a Fresnel coordinate system established along a reference line of the road, and the Fresnel coordinate system includes an arc major axis and a lateral distance axis, a sampling point on the obstacle is projected onto the reference line of the road according to the obstacle information to obtain a first Fresnel coordinate of the sampling point, and a predicted vehicle position of the vehicle at different times is projected onto the reference line of the road according to the predicted trajectory to obtain a second Fresnel coordinate of the vehicle; According to the first Freynes coordinate and the second Freynes coordinate, the transverse coordinate parameters and arc length coordinate parameters of the constraint generation boundary are determined, wherein the transverse coordinate parameters of the constraint generation boundary are expressed by transverse distance axis coordinates, and the arc length coordinate parameters of the constraint generation boundary are expressed by arc length axis coordinates.

6. The method according to claim 5, characterized in that The step of determining arc length coordinate parameters of the constraint generating boundary according to the first Fresnel coordinates and the second Fresnel coordinates includes: When the vehicle and the obstacle meet the advance avoidance condition, determining the encounter position coordinates of the vehicle and the obstacle when traveling along the predicted trajectory on the arc major axis according to the first Fresnel coordinates and the second Fresnel coordinates, wherein the encounter position coordinates are expressed in arc major axis coordinates; Determining the coordinates of the advance avoidance position on the arc major axis according to the encounter position and the preset advance avoidance distance; The starting point of the constraint generation boundary is adjusted from the encounter position coordinate to the advance avoidance position coordinate to obtain the arc length coordinate parameter of the constraint generation boundary.

7. The method according to claim 5, characterized in that Determining the transverse coordinate parameters of the constraint generation boundary according to the first Fresnel coordinate and the second Fresnel coordinate includes: Determine a transverse coordinate parameter of an initial boundary according to the first Fresnel coordinate and the second Fresnel coordinate, wherein the transverse coordinate parameter of the initial boundary is represented by a transverse distance axis coordinate; Determining whether the obstacle is a dangerous obstacle according to the obstacle information; In the case where the type of the obstacle is a dangerous obstacle, the transverse coordinate parameters of the initial boundary are adjusted in a direction away from the obstacle to obtain the transverse coordinate parameters of the constraint generation boundary.

8. A vehicle trajectory generating device, characterized in that: include: A first determination module is used to determine the range of the vehicle's passing area and the predicted trajectory of the vehicle according to the road information of the vehicle and the obstacle information of the obstacle, wherein the predicted trajectory is within the range of the passing area, and the passing area and the predicted trajectory are represented by a predetermined coordinate system, and the predetermined coordinate system is established based on the road on which the vehicle is traveling; a second determination module, configured to determine a constraint generation boundary of an obstacle avoidance constraint in the predetermined coordinate system according to the predicted trajectory and the passage area range, wherein the constraint generation boundary is a finite length line located within the passage area range; A generation module, used to generate a boundary and an obstacle avoidance cost according to the predicted trajectory and the constraint; An optimization module is used to perform joint spatiotemporal optimization on the predicted trajectory according to the obstacle avoidance cost to generate a target trajectory of the vehicle.

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the vehicle trajectory generation method according to any one of claims 1 to 7.

10. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store programs, and the processor is used to run the programs stored in the memory, wherein the vehicle trajectory generation method described in any one of claims 1 to 7 is executed when the program is run.

Images