Vehicle path planning method, domain controller and computer program product

Abstract

The invention provides a vehicle path planning method, a domain controller and a computer program product, according to the vehicle path planning method provided by the invention, boundary constraint information including a plurality of target positions in a vehicle rear structure and a vehicle front structure can be determined in combination with position information of an obstacle influencing vehicle movement; the boundary constraint information can reflect the information of the area where the target position of the vehicle is located in order to avoid collision between the vehicle and the obstacle; therefore, path planning is carried out according to the boundary constraint information of the multiple target positions of the vehicle and the cost function constructed based on the distance between the vehicle and the road reference line, so that when the vehicle runs according to the generated running path planning result, both the vehicle rear structure and the vehicle front structure do not collide with the obstacle, and the vehicle safety is improved. And the accuracy of the vehicle driving path planning result can be improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and in particular to a vehicle path planning method, a domain controller and a computer program product. Background Art

[0002] With the continuous development and improvement of artificial intelligence, sensor technology and computing power, vehicle path planning technology is also becoming more mature. By deploying radars, cameras and other sensor devices on the vehicle, the vehicle's surrounding environment can be perceived in real time, so that the real-time environmental perception data can be integrated to provide the vehicle with safe, efficient and smooth vehicle path planning results. However, when the vehicle is currently driving in a complex environment or on a curved road with a large curvature, there may be a risk of collision due to the lack of accuracy of the vehicle's path planning results, thereby affecting the safety of the vehicle's driving process. Summary of the invention

[0003] Based on this, the present invention provides a vehicle path planning method, a domain controller and a computer program product. By adopting the vehicle path planning method, the accuracy of the vehicle's driving path planning results can be improved to reduce the risk of vehicle collision when controlling or assisting the vehicle's driving based on the driving path planning results, which is beneficial to improving the safety of the vehicle's driving process.

[0004] In one aspect, the present invention provides a vehicle path planning method, the method comprising:

[0005] Obtaining the position information of obstacles that affect the movement of the vehicle;

[0006] Determine boundary constraint information of multiple target positions at the vehicle according to the position information of the obstacle; wherein the boundary constraint information is used to reflect the area information where the target positions are located in order to avoid the collision between the vehicle and the obstacle; the target positions include: a first position in the rear structure of the vehicle, and a second position in the front structure of the vehicle;

[0007] A driving path planning result of the vehicle is generated according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line.

[0008] Further, in some embodiments, determining the boundary constraint information of multiple target positions at the vehicle according to the position information of the obstacle includes:

[0009] For any of the target positions, obtaining a preset minimum safety distance corresponding to the target position;

[0010] Determine hard boundary constraint information of the target position in the Frenet coordinate system according to the position information of the obstacle in the Frenet coordinate system and the preset minimum safety distance corresponding to the target position;

[0011] The hard boundary constraint information is used to reflect the area information that can make the distance between the target position and the obstacle greater than the preset minimum safety distance.

[0012] Furthermore, in some embodiments, determining the boundary constraint information of the plurality of target positions at the vehicle according to the position information of the obstacle further includes:

[0013] For any of the target positions, obtaining a preset maximum shrinkage distance corresponding to the target position;

[0014] Determine soft boundary constraint information of the target position in the Frenet coordinate system according to the preset maximum shrinkage distance and hard boundary constraint information of the target position in the Frenet coordinate system;

[0015] Among them, the soft boundary constraint information is used to reflect the first preset range of the minimum distance between the target position and the obstacle, the starting point of the first preset range is the preset minimum safety distance, and the end point of the first preset range is the sum of the preset minimum safety distance and the preset maximum limit distance.

[0016] Furthermore, in some embodiments, determining the boundary constraint information of the plurality of target positions at the vehicle according to the position information of the obstacle further includes:

[0017] For any of the target positions, obtaining a preset maximum safety distance corresponding to the target position;

[0018] The determining of the hard boundary constraint information of the target position in the Frenet coordinate system specifically includes:

[0019] Determine hard boundary constraint information of the target position in the Frenet coordinate system according to the position information of the obstacle in the Frenet coordinate system, the preset maximum safety distance corresponding to the target position, and the preset minimum safety distance;

[0020] The hard boundary constraint information is specifically used to reflect the area information that can make the distance between the target position and the obstacle greater than the preset minimum safety distance and less than the preset maximum safety distance;

[0021] The soft boundary constraint information is specifically used to reflect the first preset range of the minimum distance between the target position and the obstacle, and the second preset range of the maximum distance between the target position and the obstacle, the starting point of the second preset range is the difference between the preset maximum safety distance and the preset maximum reduction distance, and the ending point of the second preset range is the preset maximum safety distance.

[0022] Furthermore, in some embodiments, determining the boundary constraint information of the plurality of target positions at the vehicle according to the position information of the obstacle further includes:

[0023] If a line connecting any of the target positions and the other target positions is parallel to the vehicle centerline of the vehicle, the soft boundary constraint information of the other target positions in the Frenet coordinate system is generated according to the soft boundary constraint information of any of the target positions in the Frenet coordinate system.

[0024] Further, in some implementations, generating the driving path planning result of the vehicle according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line includes:

[0025] Obtaining a cost function constructed based on a specified ordinate of a specified position at the vehicle in the Frenet coordinate system and a boundary relaxation parameter; wherein the boundary relaxation parameter is used to reflect the degree to which a soft boundary corresponding to the soft boundary constraint information deviates from a hard boundary corresponding to the hard boundary constraint information, and the boundary relaxation parameter is negatively correlated with a value of the cost function;

[0026] Solving a quadratic programming problem based on the soft boundary constraint information of the target position and the cost function to obtain the specified ordinate data of the specified position that satisfies the soft boundary constraint information of each target position and minimizes the cost function;

[0027] Based on a preset association relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position, the vertical coordinate data of the target position is calculated according to the designated vertical coordinate data of the designated position;

[0028] A driving path planning result of the vehicle is generated according to the ordinate data of the target position.

[0029] Further, in some implementations, generating the driving path planning result of the vehicle according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line includes:

[0030] Obtaining a cost function constructed based on a specified ordinate of a specified position at the vehicle in the Frenet coordinate system;

[0031] Solving a quadratic programming problem based on the soft boundary constraint information of the target position and the cost function to obtain the specified ordinate data of the specified position that satisfies the soft boundary constraint information of each target position and minimizes the cost function;

[0032] Based on a preset association relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position, the vertical coordinate data of the target position is calculated according to the designated vertical coordinate data of the designated position;

[0033] A driving path planning result of the vehicle is generated according to the ordinate data of the target position.

[0034] Further, in some embodiments, the calculating the ordinate data of the target position based on the preset association relationship between the designated ordinate of the designated position and the ordinate of the target position according to the designated ordinate data of the designated position includes:

[0035] Based on the preset association relationship, the designated vertical coordinate data, the first vertical coordinate offset and the second vertical coordinate offset of the designated position are superimposed to obtain the vertical coordinate data of the target position;

[0036] Among them, the first longitudinal coordinate offset is used to reflect the lateral offset of the target position compared to the designated position under the influence of the vehicle heading of the vehicle; the second longitudinal coordinate offset is used to reflect the lateral offset of the target position compared to the designated position under the influence of the road curvature of the road where the vehicle is located.

[0037] Further, in some embodiments, the first position in the rear structure of the vehicle includes: at least one of the center point of the rear bumper of the vehicle, the center point of the rear axle of the vehicle, and the rear corner point of the vehicle;

[0038] The second position in the front structure of the vehicle includes at least one of a center point of a front bumper of the vehicle, a center point of a front axle of the vehicle, and a front corner point of the vehicle.

[0039] On the other hand, the present invention further provides a domain controller, comprising: a processor and a memory; wherein the memory stores computer-readable instructions, and the computer-readable instructions are suitable for being loaded by the processor and executing the steps of the above method.

[0040] On the other hand, the present invention further provides a computer program product having at least one instruction stored thereon, wherein the at least one instruction implements the steps of the above method when executed by a controller.

[0041] According to the vehicle path planning method provided by the present invention, the boundary constraint information of multiple target positions in the rear structure and the front structure of the vehicle can be determined in combination with the position information of obstacles that affect the movement of the vehicle; because the boundary constraint information can reflect the area information of the target position at which the vehicle is located in order to avoid collision between the vehicle and the obstacle; therefore, path planning is performed based on the boundary constraint information of multiple target positions at the vehicle and a cost function constructed based on the distance between the vehicle and the road reference line, so that when the vehicle travels according to the generated driving path planning result, the rear structure and the front structure of the vehicle will not collide with the obstacle, which is beneficial to improving the accuracy of the vehicle driving path planning result, and is also beneficial to improving the safety of controlling or assisting the vehicle driving based on the vehicle driving path planning result.

[0042] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic diagram of a vehicle driving path planning result provided by an embodiment of the present invention;

[0044] Figure 2 A schematic diagram of a vehicle path planning method provided by an embodiment of the present invention;

[0045] Figure 3 A schematic diagram of a boundary constraint situation of a target position in a Frenet coordinate system provided by an embodiment of the present invention;

[0046] Figure 4 A schematic diagram of a vehicle driving scenario provided by an embodiment of the present invention;

[0047] Figure 5 A schematic diagram of the structure of a domain controller provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. 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 making creative work are within the scope of protection of the present invention.

[0049] In the description of one or more embodiments of the present invention, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0050] In autonomous driving and intelligent assisted driving scenarios, vehicle path planning is one of the key technologies to ensure safe and efficient vehicle driving. At present, the vehicle path planning problem can usually be transformed into a quadratic programming problem in the Frenet coordinate system. Although this vehicle path planning solution can improve solution efficiency and reduce computational complexity, it also brings some challenges. Figure 1 Provide explanation.

[0051] Figure 1 A schematic diagram of a vehicle driving path planning result provided by an embodiment of the present invention. Figure 1 As shown, assuming that there is an obstacle 12 in front of the vehicle 11, the upper boundary 102 and the lower boundary 103 of the area where the rear axle center point 101 is located can be set according to the position information of the obstacle 12 and the distance information between the rear axle center point 101 of the vehicle 11 and the obstacle 12 to avoid the collision between the vehicle 11 and the obstacle 12, so as to obtain the boundary constraint information for the vehicle 11. By solving the quadratic programming problem constructed using the boundary constraint information of the vehicle 11 and the preset cost function, the driving path planning result 104 of the vehicle 11 can be obtained.

[0052] However, when generating the driving path planning result 104 of the vehicle 11, only the relative position between the obstacle and the rear axle center point 101 of the vehicle 11 is simply considered. Therefore, when the vehicle is driving on complex road conditions or a curve with a large curvature, the front structure of the vehicle 11 may collide with the obstacle, resulting in insufficient accuracy and effectiveness of the vehicle driving path planning result, affecting the safety of the vehicle driving process.

[0053] Therefore, how to improve the accuracy and effectiveness of vehicle driving path planning results to reduce the risk of vehicle collision when controlling or assisting vehicle driving based on vehicle driving path planning results has become a technical problem that needs to be solved urgently.

[0054] Based on this, the present invention proposes a vehicle path planning method, which can combine the position information of obstacles that affect the movement of the vehicle to determine the boundary constraint information of multiple target positions in the rear structure and the front structure of the vehicle; because the boundary constraint information can reflect the area information of the target position at which the vehicle is located in order to avoid the collision between the vehicle and the obstacle; therefore, according to the boundary constraint information of the multiple target positions at the vehicle and the cost function constructed based on the distance between the vehicle and the road reference line, the path planning can be performed, so that when the vehicle travels according to the generated driving path planning results, the rear structure and the front structure of the vehicle will not collide with the obstacles, which is beneficial to improving the accuracy and effectiveness of the vehicle driving path planning results, and is also beneficial to improving the safety of controlling or assisting the vehicle driving based on the vehicle driving path planning results.

[0055] See also Figure 2 , which is a flow chart of a vehicle path planning method provided by an embodiment of the present invention. From a program perspective, the execution subject of the process may be a program for planning a driving path for a vehicle. Alternatively, the execution subject of the process may also be a domain control device or a vehicle equipped with the above program, or other devices capable of communicating with the domain control device or the vehicle, without specific limitation.

[0056] The following is for Figure 2 The process shown is described in detail, and the vehicle path planning method may specifically include the following steps:

[0057] Step S202, obtaining the position information of obstacles that affect the movement of the vehicle.

[0058] In the embodiment of the present invention, when it is necessary to plan the driving path for the vehicle, the radar equipment, image acquisition equipment or other sensors carried by the vehicle can be combined to collect the vehicle's surrounding environment information, and the target detection is performed on the vehicle's surrounding environment information to obtain the position information of obstacles that affect the movement of the vehicle. In practical applications, the obstacles can be of various types, such as traffic participants such as motor vehicles, non-motor vehicles, pedestrians, etc., or they can also be facilities such as guardrails and traffic sign poles deployed on the road, or they can also be buildings around the vehicle, etc., which are not specifically limited.

[0059] Step S204, determining boundary constraint information of multiple target positions at the vehicle based on the position information of the obstacle; wherein the boundary constraint information is used to reflect the area information of the target positions required to avoid a collision between the vehicle and the obstacle; the target positions include: a first position in the rear structure of the vehicle, and a second position in the front structure of the vehicle.

[0060] In an embodiment of the present invention, the vehicle can be divided into a vehicle rear structure and a vehicle front structure according to a preset dividing boundary. The vehicle front structure may refer to a vehicle structure including a front portion, and the vehicle rear structure may refer to a vehicle structure including a rear portion. In practical applications, the preset dividing boundary may be a straight line passing through the center point of the vehicle and parallel to the line between the center points of the left and right front wheels / the center points of the left and right rear wheels of the vehicle, or a straight line passing through the center of mass of the vehicle and parallel to the line between the center points of the left and right front wheels / the center points of the left and right rear wheels of the vehicle, or a straight line passing through the B-pillar of the vehicle and parallel to the line between the center points of the left and right front wheels / the center points of the left and right rear wheels of the vehicle, and no specific limitation is made to this.

[0061] In the embodiment of the present invention, in order to simultaneously avoid collision between the rear structure and the front structure of the vehicle and an obstacle, it is necessary to pre-select at least one first position in the rear structure of the vehicle and at least one second position in the front structure of the vehicle as target positions, so as to determine the boundary constraint information corresponding to each target position according to the position information of the obstacle and the distance information required to be maintained between the target position and the obstacle when the vehicle and the obstacle are avoided. The boundary constraint information corresponding to any target position can generally reflect the area information where the target position is located in order to avoid collision between the vehicle and the obstacle.

[0062] In practical applications, the first position in the rear structure of the vehicle may include but is not limited to: the center point of the rear axle of the vehicle, the center point of the rear bumper of the vehicle, the rear corner point of the vehicle, etc.; and the second position in the front structure of the vehicle may include but is not limited to: the center point of the front axle of the vehicle, the center point of the front bumper of the vehicle, the front corner point of the vehicle, etc., and no specific limitation is made. Among them, the rear corner point of the vehicle and the front corner point of the vehicle may be the four corner points of the rectangular area where the vehicle is currently located; and the center point of the rear bumper of the vehicle may be the center point of the side corresponding to the rear bumper in the rectangular area where the vehicle is currently located, and similarly, the center point of the front bumper of the vehicle may be the center point of the side corresponding to the front bumper in the rectangular area where the vehicle is currently located.

[0063] Step S206: Generate a driving path planning result for the vehicle according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line.

[0064] In the embodiment of the present invention, since the user's driving comfort and the safety of the vehicle's driving process are the goals of vehicle driving path planning, and when the vehicle deviates far from the road reference line (for example, the center line of the lane where the vehicle is located), the collision risk of the vehicle may increase, and when there is a large acceleration change at the vehicle, the driver and passengers will feel uncomfortable. Therefore, the cost function can be constructed in combination with the distance between the vehicle and the road reference line and the speed of change of the distance, acceleration and other parameters. By combining the boundary constraint information of multiple target positions at the vehicle, the cost function is solved, so that when the vehicle drives according to the generated driving path planning results, the rear structure and the front structure of the vehicle will not collide with obstacles, so as to ensure the accuracy of the vehicle driving path planning results and the safety of the vehicle when driving, and also ensure the comfort of the driver and passengers.

[0065] In a feasible implementation manner, determining the boundary constraint information of the plurality of target positions at the vehicle according to the position information of the obstacle may include:

[0066] For any of the target positions, a preset minimum safety distance corresponding to the target position is obtained.

[0067] According to the position information of the obstacle in the Frenet coordinate system and the preset minimum safety distance corresponding to the target position, the hard boundary constraint information of the target position in the Frenet coordinate system is determined.

[0068] The hard boundary constraint information is used to reflect the area information that can make the distance between the target position and the obstacle greater than the preset minimum safety distance.

[0069] In the embodiment of the present invention, the preset minimum safety distance corresponding to the target position may refer to: in order to avoid collision between the vehicle and the obstacle, the minimum distance between the target position of the vehicle and the obstacle is allowed to be maintained. For example, when the target position is the center point of the front / rear axle of the vehicle, the preset minimum safety distance should at least be: the lateral distance between the center point of the front / rear axle of the vehicle and the left / right side rearview mirror of the vehicle. In practical applications, in order to ensure the safety of the vehicle, the preset minimum safety distance is usually made the sum of the lateral distance between the center point of the front / rear axle of the vehicle and the left / right side rearview mirror of the vehicle and the preset distance (for example, 0.2 meters, 0.5 meters, etc.). Alternatively, when the target position is the corner point of the vehicle away from the obstacle (for example, when the obstacle is on the left side of the vehicle, the target position can be the right front corner point or the right rear corner point of the vehicle), the preset minimum safety distance should at least be: the width value of the vehicle; similarly, in order to ensure the safety of the vehicle, the preset minimum safety distance is usually made the sum of the width value of the vehicle and the preset distance (for example, 0.2 meters, 0.5 meters, etc.); no specific limitation is made on this.

[0070] In an embodiment of the present invention, hard boundary constraint information may refer to boundary constraint information that must be strictly satisfied and is fixed. Since the vehicle will not collide with the obstacle when the distance between the target position of the vehicle and the obstacle is greater than the preset minimum safety distance, and the vehicle driving path planning problem needs to be converted into a quadratic planning problem in the Frenet coordinate system, the boundary constraint information reflecting the area information in the Frenet coordinate system that can make the distance between the target position and the obstacle greater than the preset minimum safety distance can be constructed based on the position information of the obstacle in the Frenet coordinate system and the preset minimum safety distance corresponding to the target position, so as to serve as the hard boundary constraint information of the target position in the Frenet coordinate system, which is convenient and quick.

[0071] Figure 3 A schematic diagram of a boundary constraint of a target position in a Frenet coordinate system provided by an embodiment of the present invention. Figure 3 As shown, assuming that there is an obstacle 31 in front of the vehicle 32, the hard boundary constraint information of the target position of the vehicle (e.g., the center point 305 of the rear axle of the vehicle) in the Frenet coordinate system can at least be used to indicate the position information of the first upper boundary 301. The minimum distance between the first upper boundary 301 and the obstacle 31 can be the preset minimum safety distance corresponding to the target position (e.g., the center point 305 of the rear axle of the vehicle).

[0072] In practical applications, in order to avoid frequent lane changes that may interfere with vehicle driving safety and user driving comfort, the maximum distance allowed between the target position of the vehicle and the obstacle can usually be determined based on the lane boundary line information of the lane the vehicle is currently in, and used as the hard boundary constraint information of the target position. This can effectively limit the activity space of the target position of the vehicle, which is conducive to prompting the vehicle to stay in the lane, thereby improving vehicle driving safety and user driving comfort.

[0073] Based on this, determining the boundary constraint information of the multiple target positions at the vehicle according to the position information of the obstacle may also include:

[0074] For any of the target positions, a preset maximum safety distance corresponding to the target position is obtained.

[0075] Determining the hard boundary constraint information of the target position in the Frenet coordinate system may specifically include:

[0076] According to the position information of the obstacle in the Frenet coordinate system, the preset maximum safety distance and the preset minimum safety distance corresponding to the target position, the hard boundary constraint information of the target position in the Frenet coordinate system is determined.

[0077] The hard boundary constraint information is specifically used to reflect the area information that can make the distance between the target position and the obstacle greater than the preset minimum safety distance and less than the preset maximum safety distance.

[0078] In the embodiment of the present invention, the preset maximum safety distance corresponding to the target position may refer to: in order to keep the vehicle in the lane and reduce the risk of crossing multiple lanes, the maximum distance allowed between the target position of the vehicle and the obstacle.

[0079] By using the preset maximum safety distance to construct the first lower boundary of the hard boundary of the target position, and using the preset minimum safety distance to construct the first upper boundary of the hard boundary of the target position, the distance between the target position and the obstacle can be made greater than the preset minimum safety distance and less than the preset maximum safety distance, which is beneficial to ensuring the safety of the vehicle driving process.

[0080] For ease of understanding, combined Figure 3 The content in the following example illustrates the hard boundary of the target location. Figure 3As shown, assuming that there is an obstacle 31 in front of the vehicle 32, the hard boundary constraint information of the target position of the vehicle (for example, the center point 305 of the rear axle of the vehicle) in the Frenet coordinate system can at least be used to indicate the position information of hard boundaries such as the first upper boundary 301 and the first lower boundary 302. The minimum distance between the first upper boundary 301 and the obstacle 31 can be the preset minimum safety distance corresponding to the target position; the maximum distance between the first lower boundary 302 and the obstacle 31 can be the preset maximum safety distance corresponding to the target position.

[0081] At present, when the hard boundary constraint information of the target position of the vehicle is used to plan the vehicle's driving path, it may have an adverse effect due to over-reliance on the preset minimum safety distance corresponding to the target position. For example, when the preset minimum safety distance corresponding to the target position is set too small, even if there is enough space on the road where the vehicle is currently located to allow the vehicle to bypass obstacles, the final generated vehicle driving path will be closer to the obstacles, thereby increasing the risk of vehicle collision; and when the preset minimum safety distance corresponding to the target position is set too large, the area where the target position can be located will be too small, thereby bringing unnecessary restrictions to the vehicle's driving path planning process, affecting the vehicle's traffic capacity and efficiency when driving based on the driving path planning results.

[0082] Based on this, in a feasible implementation manner, determining the boundary constraint information of the multiple target positions at the vehicle according to the position information of the obstacle may also include:

[0083] For any of the target positions, a preset maximum shrinkage distance corresponding to the target position is obtained.

[0084] According to the preset maximum shrinkage distance and the hard boundary constraint information of the target position in the Frenet coordinate system, the soft boundary constraint information of the target position in the Frenet coordinate system is determined.

[0085] Among them, the soft boundary constraint information is used to reflect the first preset range of the minimum distance between the target position and the obstacle, the starting point of the first preset range is the preset minimum safety distance, and the end point of the first preset range is the sum of the preset minimum safety distance and the preset maximum limit distance.

[0086] The soft boundary constraint information can also be used to reflect a second preset range of the maximum distance between the target position and the obstacle. The starting point of the second preset range can be the difference between the preset maximum safety distance and the preset maximum reduction distance, and the ending point of the second preset range can be the preset maximum safety distance.

[0087] In an embodiment of the present invention, the preset maximum shrinkage distance corresponding to the target position may refer to: when the hard boundaries such as the first upper boundary and the first lower boundary reflected in the hard boundary constraint information of the target position are shrinkage-adjusted, the maximum distance allowed for adjusting the first upper boundary, and the maximum distance allowed for adjusting the first lower boundary. Among them, the shrinkage adjustment can be used to reduce the area where the target position can be located. In actual applications, the preset maximum shrinkage distance can be set according to actual needs, for example, it can be 0.3 meters, 0.5 meters, etc., and the preset maximum shrinkage distance should generally be less than the minimum distance between the first upper boundary and the first lower boundary, so as to avoid the problem of vehicle path planning failure. In actual applications, the preset maximum shrinkage distances corresponding to the target position at different road positions can be the same or different, and there is no specific limitation on this.

[0088] In an embodiment of the present invention, soft boundary constraint information may refer to boundary constraint information with adjustable upper and lower boundaries. When the target position of the vehicle is within the first upper boundary and the first lower boundary reflected by the hard boundary constraint information, the vehicle will not collide with the obstacle usually; however, at this time, the hard boundary constraint may cause the vehicle's driving path planning process to be less flexible, and the planned vehicle driving path may be too close to the obstacle, so there is a need to limit and adjust the first upper boundary and the first lower boundary reflected by the hard boundary constraint information. Based on this, the upper boundary and the lower boundary that are located within the area formed by the first upper boundary to the first lower boundary and can be adjusted within the size range of the above-mentioned preset maximum limit distance can be used as the second upper boundary and the second lower boundary of the target position, thereby obtaining the soft boundary reflected by the soft boundary constraint information of the target position in the Frenet coordinate system.

[0089] For ease of understanding, combined Figure 3 Explain the soft boundary constraint information of the target position of the vehicle in the Frenet coordinate system. Figure 3 As shown, it is assumed that the hard boundary constraint information of the target position at the vehicle (for example, the center point 305 of the rear axle of the vehicle) in the Frenet coordinate system is used to reflect the hard boundaries such as the first upper boundary 301 and the first lower boundary 302. In addition, the distance between the first dotted line 303 and the first upper boundary 301 is the preset maximum shrinkage distance corresponding to the target position, and the distance between the second dotted line 304 and the first lower boundary 302 can also be the preset maximum shrinkage distance corresponding to the target position, then the soft boundary constraint information of the target position in the Frenet coordinate system can reflect: the second upper boundary of the target position is between the first dotted line 303 and the first upper boundary 301, and the second lower boundary of the target position is between the second dotted line 304 and the first lower boundary 302, thereby obtaining the soft boundaries such as the second upper boundary and the second lower boundary of the target position.

[0090] Since the area where the target is allowed to be located reflected by the soft boundary constraint information of the target position (i.e., the area between the second lower boundary and the second upper boundary) is all located within the area where the target is allowed to be located reflected by the hard boundary constraint information of the target position (the area between the first lower boundary and the first upper boundary), the vehicle driving path planning result generated based on the soft boundary constraint information of the target position can effectively avoid collision between the vehicle and obstacles, which is beneficial to ensuring the accuracy and effectiveness of the vehicle driving path planning result; and, since the positions of the second upper boundary and the second lower boundary corresponding to the soft boundary constraint information of the target position can be limited and adjusted, it has good flexibility and is also beneficial to avoiding the problem of the planned vehicle driving path being too close to obstacles, and reducing the probability of failure in vehicle driving path planning.

[0091] In the embodiments of the present specification, the soft boundary constraint information of each target position can be calculated respectively according to the preset minimum safety distance, preset maximum safety distance and preset maximum shrinkage distance corresponding to each target position, with good accuracy. In practical applications, the preset minimum safety distance, preset maximum safety distance and preset maximum shrinkage distance corresponding to different target positions at the same road position can be the same or different, and no specific limitation is made. In addition, when the line between two target positions is parallel to the vehicle centerline of the vehicle, the upper and lower boundaries of the two target positions can usually be consistent, therefore, only the soft boundary constraint information of a single target position can be calculated; so as to combine the soft boundary constraint information of this target position and the relative position relationship between the above two target positions, to determine the soft boundary constraint information of another target position, which is conducive to saving computing resources.

[0092] Based on this, in a feasible implementation manner, determining the boundary constraint information of the multiple target positions at the vehicle according to the position information of the obstacle may also include:

[0093] If a line connecting any of the target positions and the other target positions is parallel to the vehicle centerline of the vehicle, the soft boundary constraint information of the other target positions in the Frenet coordinate system is generated according to the soft boundary constraint information of any of the target positions in the Frenet coordinate system.

[0094] In the embodiments of the present specification, the horizontal axis of the Frenet coordinate system can be used to reflect the longitudinal displacement along the road reference line, and the vertical axis of the Frenet coordinate system can be used to reflect the normal distance between each position and the road reference line; wherein, the road reference line can be the center line of the lane in which the vehicle is currently located, or it can also be other road boundary lines, without specific limitation.

[0095] In order to facilitate the understanding of the principle of generating soft boundary constraint information of other target positions based on the soft boundary constraint information of a certain target position, Figure 3 Explain. Figure 3 As shown, assuming that the vehicle 32 is traveling from left to right, the target position of the vehicle includes the center point 305 of the vehicle's rear axle and the center point 306 of the vehicle's front bumper. If the second upper boundary corresponding to the soft constraint information of the center point 305 of the vehicle's rear axle is located within the straight line 301 and the dotted line 303, and the second lower boundary corresponding to the soft constraint information of the center point 305 of the vehicle's rear axle is located within the straight line 302 and the dotted line 304. Since the center point 305 of the vehicle's rear axle and the center point 306 of the vehicle's front bumper can both be located on the vehicle center line, the soft constraint information of the center point 306 of the vehicle's front bumper can be determined based on the soft constraint information of the center point 305 of the vehicle's rear axle.

[0096] Specifically, the soft boundary constraint information of the center point 305 of the rear axle of the vehicle may include a preset number of boundary points in the Frenet coordinate system (i.e., the discrete points constituting the second upper boundary and the second lower boundary). If the lateral distances of the above boundary points in the Frenet coordinate system are all preset constraint steps, and the difference between the horizontal coordinates of the center point 305 of the rear axle of the vehicle and the center point 306 of the front bumper of the vehicle in the Frenet coordinate system is the target difference, then the i-th boundary point in the soft boundary constraint information of the center point 306 of the front bumper of the vehicle may be the j-th boundary point in the soft boundary constraint information of the center point 305 of the rear axle of the vehicle, where j is the sum of the integer value obtained by rounding up the quotient of the target difference and the preset constraint step and i, which will not be elaborated on.

[0097] In a feasible implementation manner, generating the driving path planning result of the vehicle according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line may include:

[0098] Obtain a cost function constructed based on the specified ordinate of the specified position at the vehicle in the Frenet coordinate system and a boundary relaxation parameter; wherein the boundary relaxation parameter can be used to reflect the degree to which the soft boundary corresponding to the soft boundary constraint information deviates from the hard boundary corresponding to the hard boundary constraint information, and the boundary relaxation parameter is negatively correlated with the value of the cost function.

[0099] A quadratic programming problem constructed based on the soft boundary constraint information of the target position and the cost function is solved to obtain the specified vertical coordinate data of the specified position that satisfies the soft boundary constraint information of each target position and minimizes the cost function.

[0100] Based on a preset association relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position, the vertical coordinate data of the target position is calculated according to the designated vertical coordinate data of the designated position.

[0101] A driving path planning result of the vehicle is generated according to the ordinate data of the target position.

[0102] In the present embodiment, since keeping the vehicle centered in the lane can be used as a path planning goal when planning the vehicle's driving path, the smaller the designated vertical coordinate of the designated position of the vehicle in the Frenet coordinate system, the smaller the degree of deviation of the vehicle from the center of the lane, and the smoother and safer the vehicle's operation, the better. Therefore, the designated vertical coordinate of the designated position can be combined to build a cost function, and the designated vertical coordinate data of the designated position can be positively correlated with the value of the cost function. The designated position can be set according to actual needs. For example, the designated position can be the center point of the rear axle of the vehicle, the center of mass of the vehicle, the center point of the front axle of the vehicle, the center point of the front bumper of the vehicle, etc., and no specific limitation is made to this.

[0103] In the present illustrative embodiment, the soft boundary constraint information of any target position usually includes a plurality of soft boundary points set along the extension direction of the road where the vehicle is located, thereby forming soft boundaries such as the second upper boundary and the second lower boundary corresponding to the soft boundary constraint information. Since when the soft boundary point deviates from the corresponding hard boundary by a large distance, the vehicle driving path finally planned will usually be farther away from obstacles, and the vehicle is more likely to stay in the lane, which is beneficial to improving the safety of vehicle operation. Based on this, a boundary relaxation parameter can be set to reflect the degree (for example, distance) of the soft boundary corresponding to the soft boundary constraint information deviating from the hard boundary corresponding to the hard boundary constraint information, so as to build a cost function in combination with the boundary relaxation parameter, and the boundary relaxation parameter and the value of the cost function can also be negatively correlated.

[0104] In an embodiment of the present specification, when solving a quadratic programming problem based on the soft boundary constraint information of each target position and the cost function, it is necessary to minimize the function value of the cost function on the basis of satisfying the soft boundary constraint information of each target position; and, since the specified vertical coordinate of the specified position in the cost function can be an optimization quantity, solving the above-mentioned quadratic programming problem can obtain the specified vertical coordinate data of the specified position in the Frenet coordinate system when satisfying the soft boundary constraint information of each target position and minimizing the cost function.

[0105] In a feasible implementation manner, generating the driving path planning result of the vehicle according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line may include:

[0106] A cost function constructed based on a specified ordinate of a specified position at the vehicle in the Frenet coordinate system is obtained.

[0107] A quadratic programming problem constructed based on the soft boundary constraint information of the target position and the cost function is solved to obtain the specified vertical coordinate data of the specified position that satisfies the soft boundary constraint information of each target position and minimizes the cost function.

[0108] Based on a preset association relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position, the vertical coordinate data of the target position is calculated according to the designated vertical coordinate data of the designated position.

[0109] A driving path planning result of the vehicle is generated according to the ordinate data of the target position.

[0110] In the embodiment of this specification, the cost function can also be constructed in combination with the specified ordinate of the specified position at the vehicle, without combining the boundary relaxation parameter to construct the cost function, which has good flexibility. Since the principle of subsequent vehicle driving path planning based on the cost function and the soft boundary constraint information of the target position at the vehicle can be consistent with the vehicle driving path planning principle given in the above embodiment, it will not be repeated.

[0111] It is understandable that, due to the large curvature of the curve, if the boundary of the front structure of the vehicle is over-constrained, the vehicle's driving path may not be successfully planned. Therefore, under certain conditions, the front structure of the vehicle may not satisfy the hard boundary constraint information to ensure that the vehicle's driving path can be successfully planned. Subsequently, the actual boundary constraint information of the target position at the front structure of the vehicle, the soft boundary constraint information of the target position at the rear structure of the vehicle, and the above-mentioned cost function can be combined to construct a quadratic programming problem, so as to solve the quadratic programming problem and obtain the specified vertical coordinate data of the specified position in the Frenet coordinate system when the boundary constraint information is satisfied and the cost function is minimized. This is conducive to improving the success rate of the vehicle's driving path specification, and thus can improve the vehicle's traffic performance.

[0112] In practical applications, since it is necessary to combine the coordinate data of each target position at the vehicle, the vehicle speed and other relevant data to generate the final vehicle driving path planning result, based on this, it is also necessary to convert and calculate the specified vertical coordinate data of the specified position according to the preset association relationship between the specified position at the vehicle and the vertical coordinates of each target position to obtain the vertical coordinate data of the target position, and then generate the vehicle driving path planning result, which has good flexibility.

[0113] In a feasible implementation manner, the calculating the ordinate data of the target position based on the preset association relationship between the designated ordinate of the designated position and the ordinate of the target position according to the designated ordinate data of the designated position may include:

[0114] Based on the preset association relationship, the designated vertical coordinate data, the first vertical coordinate offset and the second vertical coordinate offset of the designated position are superimposed to obtain the vertical coordinate data of the target position.

[0115] Among them, the first longitudinal coordinate offset can be used to reflect the lateral offset of the target position compared to the designated position under the influence of the vehicle heading of the vehicle; the second longitudinal coordinate offset can be used to reflect the lateral offset of the target position compared to the designated position under the influence of the road curvature of the road on which the vehicle is located.

[0116] In the embodiment of the present invention, in order to facilitate understanding of the preset association relationship between the specified ordinate of the specified position of the vehicle and the ordinate of the target position of the vehicle, Figure 4 Give an example. Figure 4 A schematic diagram of a vehicle driving scene provided by an embodiment of the present invention; Figure 4 As shown, it is assumed that, in the Cartesian coordinate system, the designated position 402 at the vehicle 401 is the center point of the rear axle of the vehicle, the target position 403 of the vehicle is the center point of the front bumper of the vehicle, and the curve with an arrow is the road reference line 404 used to construct the horizontal axis of the Frenet coordinate system. Then, the projection point 405 of the designated position 402 at the road reference line 404 can be determined first; a tangent line of the road reference line 404 is made through the projection point 405, then the mapping point 406 corresponding to the target position 403 can be located on the tangent line, and the distance between the mapping point 406 corresponding to the target position 403 and the projection point 405 corresponding to the designated position 402 is equal to the distance between the target position 403 and the designated position 402. At this time, the dotted rectangular area 407 containing the above-mentioned mapping point 406 and the projection point 405 is equivalent to the area where the vehicle is located when the designated position of the vehicle is located on the road reference line and the vehicle heading is tangent to the road reference line.

[0117] Since the curvature of the road reference line 404 is not 0, the distances between the projection point 405 (equivalent to the designated position of the vehicle) and the mapping point 406 (equivalent to the target position of the vehicle) in the dotted rectangular area 407 and the road reference line 404 are not the same, that is, under the influence of the road curvature of the road on which the vehicle is located, the target position of the vehicle has a lateral offset compared to the designated position in the Frenet coordinate system. Therefore, the preset correlation relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position can be reflected: on the basis of the designated vertical coordinate of the designated position, a second vertical coordinate offset reflecting the lateral offset of the target position compared to the designated position under the influence of the road curvature of the road on which the vehicle is located needs to be superimposed to obtain the vertical coordinate of the target position.

[0118] In addition, when the vehicle heading is not parallel to the tangent direction of the road reference line, for example, the position of the fixed projection point 405 is not moved, and the line between the projection point 405 and the mapping point 406 is rotated counterclockwise, at this time, the distance between the projection point 405 (equivalent to the designated position at the vehicle) and the road reference line 404 does not change, but the distance between the mapping point 406 (equivalent to the target position at the vehicle) and the road reference line 404 changes. It can be seen that under the influence of the vehicle heading, the target position of the vehicle has a lateral offset compared to the designated position in the Frenet coordinate system. Therefore, the preset association relationship between the designated ordinate of the designated position and the ordinate of the target position can be made to reflect: on the basis of the designated ordinate of the designated position, a first ordinate offset is also required to be superimposed to reflect the lateral offset of the target position compared to the designated position under the influence of the vehicle heading of the vehicle, so as to obtain the ordinate of the target position.

[0119] In actual applications, as the vehicle heading, road curvature, vehicle location on the left or right side of the road reference line, and target location on the left or right side of the specified location are different, the first ordinate offset and the second ordinate offset may be positive, negative, or even 0 to ensure the accuracy of the generated ordinate data of the target location, which will not be elaborated here.

[0120] In one embodiment, the present invention also provides Figure 5 The structural diagram of the domain controller is shown in Figure 1. Figure 5At the hardware level, the domain controller may include a processor 51 and a memory 55, and may also include an internal bus 52, a network interface 53, a memory 54, and other hardware required for other services. The domain controller may be set in a vehicle, wherein the processor 51 may read the corresponding computer-readable instructions from the memory 55 into the memory and then run them to implement the above-mentioned vehicle path planning method. The specific execution process may refer to the specific descriptions in the above-mentioned embodiments, and will not be repeated here.

[0121] In one embodiment, the present invention further provides a computer program product having at least one instruction stored thereon, and when the at least one instruction is executed by a controller, the steps of the above-mentioned vehicle path planning method are implemented. The specific execution process can be referred to the specific description in the above-mentioned embodiments, which will not be repeated here.

[0122] Finally, each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments such as the domain controller and the computer program product, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0123] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A vehicle path planning method, comprising: Obtaining the position information of obstacles that affect the movement of the vehicle; Determine boundary constraint information of multiple target positions at the vehicle according to the position information of the obstacle; wherein the boundary constraint information is used to reflect the area information where the target positions are located in order to avoid the collision between the vehicle and the obstacle; the target positions include: a first position in the rear structure of the vehicle, and a second position in the front structure of the vehicle; A driving path planning result of the vehicle is generated according to the boundary constraint information of the multiple target positions and a cost function constructed based on the distance between the vehicle and the road reference line.

2. The method according to claim 1, wherein determining boundary constraint information of multiple target positions at the vehicle according to the position information of the obstacle comprises: For any of the target positions, obtaining a preset minimum safety distance corresponding to the target position; Determine hard boundary constraint information of the target position in the Frenet coordinate system according to the position information of the obstacle in the Frenet coordinate system and the preset minimum safety distance corresponding to the target position; The hard boundary constraint information is used to reflect the area information that can make the distance between the target position and the obstacle greater than the preset minimum safety distance.

3. The method according to claim 2, wherein determining boundary constraint information of a plurality of target positions at the vehicle according to the position information of the obstacle further comprises: For any of the target positions, obtaining a preset maximum shrinkage distance corresponding to the target position; Determine soft boundary constraint information of the target position in the Frenet coordinate system according to the preset maximum shrinkage distance and hard boundary constraint information of the target position in the Frenet coordinate system; Among them, the soft boundary constraint information is used to reflect the first preset range of the minimum distance between the target position and the obstacle, the starting point of the first preset range is the preset minimum safety distance, and the end point of the first preset range is the sum of the preset minimum safety distance and the preset maximum limit distance.

4. The method according to claim 3, wherein determining the boundary constraint information of the plurality of target positions at the vehicle according to the position information of the obstacle further comprises: For any of the target positions, obtaining a preset maximum safety distance corresponding to the target position; The determining of the hard boundary constraint information of the target position in the Frenet coordinate system specifically includes: Determine hard boundary constraint information of the target position in the Frenet coordinate system according to the position information of the obstacle in the Frenet coordinate system, the preset maximum safety distance corresponding to the target position, and the preset minimum safety distance; The hard boundary constraint information is specifically used to reflect the area information that can make the distance between the target position and the obstacle greater than the preset minimum safety distance and less than the preset maximum safety distance; The soft boundary constraint information is specifically used to reflect the first preset range of the minimum distance between the target position and the obstacle, and the second preset range of the maximum distance between the target position and the obstacle, the starting point of the second preset range is the difference between the preset maximum safety distance and the preset maximum reduction distance, and the ending point of the second preset range is the preset maximum safety distance.

5. The method according to claim 4, wherein determining boundary constraint information of a plurality of target positions at the vehicle according to the position information of the obstacle further comprises: If a line connecting any of the target positions and the other target positions is parallel to the vehicle centerline of the vehicle, the soft boundary constraint information of the other target positions in the Frenet coordinate system is generated according to the soft boundary constraint information of any of the target positions in the Frenet coordinate system.

6. The method according to claim 4, wherein the generating the driving path planning result of the vehicle based on the boundary constraint information of the plurality of target positions and the cost function constructed based on the distance between the vehicle and the road reference line comprises: Obtaining a cost function constructed based on a specified ordinate of a specified position at the vehicle in the Frenet coordinate system and a boundary relaxation parameter; wherein the boundary relaxation parameter is used to reflect the degree to which a soft boundary corresponding to the soft boundary constraint information deviates from a hard boundary corresponding to the hard boundary constraint information, and the boundary relaxation parameter is negatively correlated with a value of the cost function; Solving a quadratic programming problem based on the soft boundary constraint information of the target position and the cost function to obtain the specified ordinate data of the specified position that satisfies the soft boundary constraint information of each target position and minimizes the cost function; Based on a preset association relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position, the vertical coordinate data of the target position is calculated according to the designated vertical coordinate data of the designated position; A driving path planning result of the vehicle is generated according to the ordinate data of the target position.

7. The method according to claim 4, wherein the generating the driving path planning result of the vehicle based on the boundary constraint information of the plurality of target positions and the cost function constructed based on the distance between the vehicle and the road reference line comprises: Obtaining a cost function constructed based on a specified ordinate of a specified position at the vehicle in the Frenet coordinate system; Solving a quadratic programming problem based on the soft boundary constraint information of the target position and the cost function to obtain the specified ordinate data of the specified position that satisfies the soft boundary constraint information of each target position and minimizes the cost function; Based on a preset association relationship between the designated vertical coordinate of the designated position and the vertical coordinate of the target position, the vertical coordinate data of the target position is calculated according to the designated vertical coordinate data of the designated position; A driving path planning result of the vehicle is generated according to the ordinate data of the target position.

8. The method according to claim 6 or 7, wherein the ordinate data of the target position is calculated based on the preset association relationship between the designated ordinate of the designated position and the ordinate of the target position, and according to the designated ordinate data of the designated position, comprises: Based on the preset association relationship, the designated vertical coordinate data, the first vertical coordinate offset and the second vertical coordinate offset of the designated position are superimposed to obtain the vertical coordinate data of the target position; Among them, the first longitudinal coordinate offset is used to reflect the lateral offset of the target position compared to the designated position under the influence of the vehicle heading of the vehicle; the second longitudinal coordinate offset is used to reflect the lateral offset of the target position compared to the designated position under the influence of the road curvature of the road where the vehicle is located.

9. The method according to any one of claims 1 to 7, wherein the first position in the rear structure of the vehicle comprises: At least one of the center point of the rear bumper of the vehicle, the center point of the rear axle of the vehicle and the rear corner point of the vehicle; The second position in the front structure of the vehicle includes at least one of a center point of a front bumper of the vehicle, a center point of a front axle of the vehicle, and a front corner point of the vehicle.

10. A domain controller, comprising: A processor and a memory; wherein the memory stores computer-readable instructions, and the computer-readable instructions are suitable for being loaded by the processor and executing the steps of the method as claimed in any one of claims 1 to 9.

11. A computer program product having at least one instruction stored thereon, wherein when the at least one instruction is executed by a controller, the steps of the method according to any one of claims 1 to 9 are implemented.

Images