Path planning method and device for automatic driving and electronic equipment

Abstract

The invention provides an automatic driving path planning method and device, electronic equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: establishing a solution space of an obstacle avoidance path according to obstacle information in a lane where a vehicle is located and an adjacent lane; performing sampling processing in the solution space to obtain a plurality of discrete road points, and planning a plurality of obstacle avoidance paths according to the plurality of discrete road points; wherein each obstacle avoidance path takes the position of the own vehicle as a starting point and takes the farthest discrete road point as a terminal point, and the farthest discrete road point represents the discrete road point with the longest longitudinal distance from the own vehicle in the plurality of discrete road points; calculating driving costs respectively corresponding to the plurality of obstacle avoidance paths; screening the plurality of obstacle avoidance paths according to the driving costs corresponding to the plurality of obstacle avoidance paths to obtain a target obstacle avoidance path; and controlling the vehicle to run according to the target obstacle avoidance path. According to the invention, the driving efficiency can be improved on the premise of ensuring the driving safety.

Description

Technical Field

[0001] The present application relates to vehicle technology, and in particular to a path planning method, device, electronic device, computer-readable storage medium and computer program product for autonomous driving. Background Art

[0002] Autonomous driving technology, also known as unmanned driving technology, is a type of intelligent connected vehicle technology that uses computer systems to achieve unmanned driving. Autonomous driving technology relies on the collaboration of multiple advanced technologies, such as artificial intelligence, visual computing, radar, and global positioning systems, to enable vehicles to drive automatically and safely without human active operation.

[0003] Autonomous obstacle avoidance is a very important part of autonomous driving technology, and has a direct impact on the intelligence and safety of autonomous driving technology. In the solutions provided by related technologies, obstacle avoidance is usually achieved in the own lane (i.e. the lane where the vehicle is located). However, the obstacle avoidance path planned in this way may affect the driving efficiency of the vehicle. Summary of the invention

[0004] The present application provides a path planning method, device, electronic device, computer-readable storage medium and computer program product for autonomous driving, which can improve driving efficiency while ensuring driving safety.

[0005] The technical solution of this application is implemented as follows:

[0006] The present application provides a path planning method for autonomous driving, comprising:

[0007] Establish a solution space for obstacle avoidance paths based on the obstacle information in the lane where the vehicle is located and in adjacent lanes;

[0008] Sampling is performed in the solution space to obtain a plurality of discrete waypoints, and a plurality of obstacle avoidance paths are planned according to the plurality of discrete waypoints; wherein each obstacle avoidance path starts from the position of the ego vehicle and ends at the most distant discrete waypoint, and the most distant discrete waypoint represents a discrete waypoint with the largest longitudinal distance from the ego vehicle among the plurality of discrete waypoints;

[0009] Calculating the driving costs corresponding to the plurality of obstacle avoidance paths respectively;

[0010] Filtering the plurality of obstacle avoidance paths according to the driving costs respectively corresponding to the plurality of obstacle avoidance paths to obtain a target obstacle avoidance path;

[0011] The vehicle is controlled to travel according to the target obstacle avoidance path.

[0012] The present application provides a path planning device for autonomous driving, comprising:

[0013] Establish a module for establishing a solution space for an obstacle avoidance path based on obstacle information in the lane where the vehicle is located and in adjacent lanes;

[0014] A planning module, configured to perform sampling processing in the solution space to obtain a plurality of discrete waypoints, and plan a plurality of obstacle avoidance paths according to the plurality of discrete waypoints; wherein each obstacle avoidance path starts from the position of the ego vehicle and ends at the most distant discrete waypoint, wherein the most distant discrete waypoint represents a discrete waypoint with the largest longitudinal distance from the ego vehicle among the plurality of discrete waypoints;

[0015] A calculation module, used to calculate the driving costs corresponding to the plurality of obstacle avoidance paths respectively;

[0016] A screening module, used for screening the plurality of obstacle avoidance paths according to the driving costs respectively corresponding to the plurality of obstacle avoidance paths, to obtain a target obstacle avoidance path;

[0017] A control module is used to control the vehicle to travel according to the target obstacle avoidance path.

[0018] The present application provides an electronic device, including:

[0019] A memory for storing executable instructions;

[0020] The processor is used to implement the path planning method for autonomous driving provided in the present application when executing the executable instructions stored in the memory.

[0021] The present application provides a computer-readable storage medium storing executable instructions for causing a processor to execute and implement the path planning method for autonomous driving provided in the present application.

[0022] The present application provides a computer program product, which includes executable instructions for causing a processor to execute and implement the path planning method for autonomous driving provided in the present application.

[0023] This application has the following beneficial effects:

[0024] This application establishes a solution space for obstacle avoidance paths based on obstacle information in the lane where the vehicle is located and in adjacent lanes; performs sampling processing in the solution space to obtain multiple discrete waypoints, and plans multiple obstacle avoidance paths based on the multiple discrete waypoints; wherein each obstacle avoidance path starts from the position of the vehicle and ends at the farthest scattered waypoint, and the farthest scattered waypoint represents the discrete waypoint with the largest longitudinal distance from the vehicle among the multiple discrete waypoints; calculates the driving costs corresponding to the multiple obstacle avoidance paths; screens the multiple obstacle avoidance paths according to the driving costs corresponding to the multiple obstacle avoidance paths to obtain the target obstacle avoidance path; controls the vehicle to drive along the target obstacle avoidance path. This application comprehensively considers multiple lanes rather than being limited to the current lane, making it easy to make decisions on whether to avoid obstacles in the current lane or to change lanes, and ultimately obtains the target obstacle avoidance path with the lowest driving cost, which can maximize driving efficiency while ensuring driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 This is a schematic diagram of the architecture of the path planning system for autonomous driving provided in an embodiment of the present application;

[0027] Figure 2 is a structural schematic diagram of a vehicle-mounted device provided in an embodiment of the present application;

[0028] Figure 3A This is a first flow chart of the path planning method for autonomous driving provided in an embodiment of the present application;

[0029] Figure 3B This is a second flow chart of the path planning method for autonomous driving provided in an embodiment of the present application;

[0030] Figure 3C This is a third flow chart of the path planning method for autonomous driving provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0032] In the following description, reference is made to "some embodiments", which describes a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict. In the following description, reference is made to the term "multiple" to mean at least two.

[0033] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0035] The embodiments of the present application provide a method, device, electronic device, computer-readable storage medium, and computer program product for autonomous driving path planning, which can improve driving efficiency while ensuring driving safety. The following describes an exemplary application of the electronic device provided by the embodiments of the present application. The electronic device provided by the embodiments of the present application can be implemented as a vehicle-mounted device or as a server.

[0036] See also Figure 1 , Figure 1 It is an architectural diagram of an autonomous driving path planning system 100 provided in an embodiment of the present application, wherein a vehicle-mounted device 400 is connected to a server 200 via a network 300, wherein the network 300 may be a wide area network or a local area network, or a combination of the two.

[0037] In some embodiments, taking the electronic device as a vehicle-mounted device as an example, the path planning method for autonomous driving provided in the embodiments of the present application can be implemented by the vehicle-mounted device. For example, the vehicle-mounted device 400 establishes a solution space for obstacle avoidance paths based on the obstacle information in the lane where the vehicle (referring to the vehicle deployed by the vehicle-mounted device 400) is located and in the adjacent lanes; performs sampling processing in the solution space to obtain multiple discrete waypoints, and plans multiple obstacle avoidance paths based on the multiple discrete waypoints; wherein each obstacle avoidance path starts from the position of the vehicle and ends at the farthest scattered waypoint, and the farthest scattered waypoint represents the discrete waypoint with the largest longitudinal distance from the vehicle among the multiple discrete waypoints; calculates the driving costs corresponding to the multiple obstacle avoidance paths; screens the multiple obstacle avoidance paths according to the driving costs corresponding to the multiple obstacle avoidance paths to obtain the target obstacle avoidance path; and controls the vehicle to travel according to the target obstacle avoidance path. In the above method, the vehicle-mounted device 400 implements path planning locally without transmitting data to a remote server for processing, which can greatly reduce the time and delay of data transmission, thereby improving the real-time processing and improving the response speed; at the same time, it is not restricted by network conditions, which is especially important for scenarios where the network environment is unstable or cannot be connected to the network, such as when a vehicle is traveling in a remote mountainous area without a network.

[0038] In some embodiments, the path planning method for autonomous driving provided in the embodiments of the present application can be implemented by the vehicle-mounted device and the server in collaboration. For example, the vehicle-mounted device 400 sends the obstacle information in the lane where the vehicle is located and the adjacent lane to the server 200; the server 200 establishes a solution space for the obstacle avoidance path based on the obstacle information in the lane where the vehicle is located and the adjacent lane; the server 200 performs sampling processing in the solution space to obtain multiple discrete waypoints, and plans multiple obstacle avoidance paths based on the multiple discrete waypoints; wherein each obstacle avoidance path starts from the position of the vehicle and ends at the farthest scattered waypoint, and the farthest scattered waypoint represents the discrete waypoint with the largest longitudinal distance from the vehicle among the multiple discrete waypoints; the server 200 calculates the driving costs corresponding to the multiple obstacle avoidance paths; the server 200 screens the multiple obstacle avoidance paths according to the driving costs corresponding to the multiple obstacle avoidance paths, and obtains the target obstacle avoidance path; the server 200 sends the target obstacle avoidance path to the vehicle-mounted device 400, so that the vehicle-mounted device 400 controls the vehicle to travel according to the target obstacle avoidance path. In the above method, the server 200 usually has powerful computing power and storage resources, and can efficiently implement path planning; the vehicle-mounted device 400 is mainly responsible for data collection, data transmission and control, and does not need to undertake complex computing tasks, which can reduce the local pressure of the vehicle-mounted device 400.

[0039] Take the electronic device provided in the embodiment of the present application as an example, which is a vehicle-mounted device. Figure 2 , Figure 2is a structural diagram of a vehicle-mounted device 400 provided in an embodiment of the present application, Figure 2 The vehicle-mounted device 400 shown includes: at least one processor 410, a memory 450, at least one network interface 420 and a user interface 430. The various components in the vehicle-mounted device 400 are coupled together via a bus system 440. It is understood that the bus system 440 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 440 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the bus system 440 is not described in detail. Figure 2 Various buses are labeled as bus system 440 .

[0040] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0041] The user interface 430 includes one or more output devices 431 that enable presentation of media content, including one or more speakers and / or one or more visual display screens. The user interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as microphones, touch screen displays, cameras, other input buttons and controls.

[0042] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices that are physically remote from the processor 410.

[0043] The memory 450 includes a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), and the volatile memory may be a random access memory (RAM). The memory 450 described in the embodiments of the present application is intended to include any suitable type of memory.

[0044] In some embodiments, memory 450 can store data to support various operations, examples of which include programs, modules, and data structures, or a subset or superset thereof, as exemplarily described below.

[0045] Operating system 451, including system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks;

[0046] A network communication module 452, used to reach other electronic devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 include: Bluetooth, Wireless Compatibility Certification (WiFi), and Universal Serial Bus (USB), etc.;

[0047] a presentation module 453 for enabling presentation of information via one or more output devices 431 (e.g., display screen, speaker, etc.) associated with the user interface 430 (e.g., a user interface for operating peripherals and displaying content and information);

[0048] The input processing module 454 is used to detect one or more user inputs or interactions from one of the one or more input devices 432 and translate the detected inputs or interactions.

[0049] In some embodiments, the path planning device for autonomous driving provided in the embodiments of the present application may be implemented in software. Figure 2 The path planning device 455 for autonomous driving stored in the memory 450 is shown, which can be software in the form of programs and plug-ins, including the following software modules: a building module 4551, a planning module 4552, a calculation module 4553, a screening module 4554, and a control module 4555. These modules are logical, and therefore can be arbitrarily combined or further split according to the functions implemented. The functions of each module will be described below.

[0050] The path planning method for autonomous driving provided in the embodiments of the present application will be explained in combination with the exemplary application and implementation of the electronic device provided in the embodiments of the present application.

[0051] See also Figure 3A , Figure 3A is a flow chart of the path planning method for autonomous driving provided in the embodiment of the present application, which will be combined with Figure 3A The steps shown are explained.

[0052] In step 101, a solution space for an obstacle avoidance path is established based on obstacle information in the lane where the vehicle is located and in adjacent lanes.

[0053] Here, obstacle information in the lane where the vehicle is located and the adjacent lanes is obtained, and a solution space for the obstacle avoidance path is established based on the obstacle information in the lane where the vehicle is located and the adjacent lanes, wherein the solution space can be understood as a space in the lane where the vehicle is located and the adjacent lanes that is free of obstacles.

[0054] It is worth noting that the embodiment of the present application does not limit the method of obtaining obstacle information, and the obstacle information can be obtained through the vehicle-mounted sensor of the vehicle, wherein the vehicle-mounted sensor includes at least one of a vehicle-mounted camera and a vehicle-mounted radar. For example, data can be collected through the vehicle-mounted camera and the vehicle-mounted radar respectively, and then the collected data can be fused by sensors to obtain obstacle information in the lane where the vehicle is located and the adjacent lanes, thereby improving the accuracy of the obtained obstacle information through sensor fusion.

[0055] It is worth noting that the embodiment of the present application does not limit the number of adjacent lanes. For example, the adjacent lanes may include at least one of the i lanes on the left side and the j lanes on the right side of the lane where the vehicle is located, where i and j are both integers greater than 0. For example, the lanes on a road are lane 1, lane 2, ... lane 5 from left to right. Assuming that the lane where the vehicle is located is lane 3, the adjacent lanes may include lane 2 and lane 4, or may include lane 1, lane 2, lane 4 and lane 5.

[0056] It is worth noting that the obstacle information includes at least the obstacle position and obstacle speed, and may also include the obstacle length and obstacle width.

[0057] In some embodiments, before establishing a solution space for an obstacle avoidance path based on obstacle information in the lane where the vehicle is located and in adjacent lanes, the path planning method for autonomous driving also includes: determining a first longitudinal distance based on the speed of the vehicle and a first preset time duration; obtaining obstacle information of obstacles in the lane where the vehicle is located and in adjacent lanes, and whose longitudinal distances to the vehicle are less than the first longitudinal distance.

[0058] For example, the speed of the vehicle can be multiplied by the first preset time to obtain the first longitudinal distance, and the obstacle information of obstacles in the lane where the vehicle is located and the adjacent lanes whose longitudinal distance to the vehicle is less than the first longitudinal distance can be obtained, so as to establish a solution space for the obstacle avoidance path based on the obtained obstacle information. The solution space here can be understood as the space in the lane where the vehicle is located and the adjacent lanes that is within the first longitudinal distance (with the position of the vehicle as the starting point) and has no obstacles. The above-mentioned first preset time can be set according to the actual application scenario, such as 6 to 8 seconds.

[0059] The above method only obtains obstacle information of obstacles whose longitudinal distance from the vehicle is less than the first longitudinal distance, which can ensure the necessity and accuracy of the obtained obstacle information, help reduce subsequent calculation costs, and improve path planning efficiency. In addition, by establishing a solution space within the first longitudinal distance starting from the position of the vehicle, it is possible to avoid increasing the complexity and uncertainty of path planning due to an excessively large solution space, which helps to improve the stability of path planning.

[0060] It is worth noting that the longitudinal direction involved in the embodiments of the present application refers to the direction along the lane, and the lateral direction refers to the direction perpendicular to the lane.

[0061] In step 102, sampling processing is performed in the solution space to obtain multiple discrete waypoints, and multiple obstacle avoidance paths are planned based on the multiple discrete waypoints; wherein each obstacle avoidance path starts from the position of the ego vehicle and ends at the farthest scattered waypoint, and the farthest scattered waypoint represents the discrete waypoint with the largest longitudinal distance from the ego vehicle among the multiple discrete waypoints.

[0062] Here, sampling is performed in the solution space to obtain multiple discrete waypoints, wherein the sampling process can be irregular sampling or sampling according to a specific rule, and the latter method can improve the orderliness of path planning and reduce the computational cost. Then, multiple obstacle avoidance paths are planned based on the multiple discrete waypoints, each obstacle avoidance path starts from the position of the vehicle and ends at the farthest scattered waypoint. The farthest scattered waypoint indicates the discrete waypoint with the largest longitudinal distance from the vehicle among the multiple discrete waypoints, wherein the number of the farthest scattered waypoints may be one or more.

[0063] In some embodiments, multiple obstacle avoidance paths may be planned in an exhaustive manner to avoid missing possible obstacle avoidance paths.

[0064] In some embodiments, the above-mentioned sampling processing in the solution space to obtain multiple discrete waypoints can be achieved in the following manner: sampling processing is performed in the longitudinal direction of the solution space according to the sampling longitudinal distance, and in the transverse direction of the solution space according to the sampling transverse distance, to obtain multiple discrete waypoints.

[0065] Here, regular sampling processing can be performed in the solution space to obtain multiple discrete waypoints. For example, sampling processing is performed in the longitudinal direction of the solution space according to the sampled longitudinal distance, and in the transverse direction of the solution space according to the sampled transverse distance, to obtain multiple discrete waypoints, wherein the sampled longitudinal distance and the sampled transverse distance can be preset, for example, the sampled longitudinal distance is set to 5 meters, and the sampled transverse distance is set to 0.2 meters. The above method can improve the orderliness of path planning through regular sampling processing.

[0066] For example, multiple discrete waypoints obtained through sampling processing are arranged in the form of M rows and N columns, where the row direction refers to the horizontal direction, the column direction refers to the vertical direction, and M and N are both integers greater than 1. In the planning process, starting from the position of the vehicle, a discrete waypoint is selected in each of the M rows and connected in sequence, and finally a planned obstacle avoidance path is obtained. In this way, the effectiveness of the planned obstacle avoidance path can be guaranteed, that is, it conforms to the actual driving situation.

[0067] In some embodiments, before performing sampling processing in the longitudinal direction within the solution space according to the sampled longitudinal distance and in the lateral direction within the solution space according to the sampled lateral distance, the path planning method for autonomous driving also includes: determining the second longitudinal distance based on the speed of the vehicle and a second preset time length; determining the maximum value of the second longitudinal distance and the preset longitudinal distance as the sampled longitudinal distance for sampling processing in the longitudinal direction.

[0068] Here, the second longitudinal distance is determined according to the speed of the vehicle and the second preset time, wherein the second preset time can be set according to the actual application scenario, such as being set to 0.5 seconds. Then, the maximum value of the second longitudinal distance and the preset longitudinal distance (such as 5 meters) is determined as the sampling longitudinal distance for sampling processing in the longitudinal direction. Through the above method, the rationality of the sampling longitudinal distance can be improved, and too many discrete waypoints caused by too small sampling longitudinal distance can be avoided. At the same time, it can ensure that the speed of the vehicle can support driving according to the planned obstacle avoidance path.

[0069] In step 103, the driving costs corresponding to the plurality of obstacle avoidance paths are calculated.

[0070] Here, for each planned obstacle avoidance path, the driving cost corresponding to each obstacle avoidance path is calculated, and the driving cost can directly reflect the effect of the obstacle avoidance path. The embodiment of the present application does not limit the calculation method of the driving cost, and factors such as safety and riding experience can be considered. For example, the driving cost corresponding to each obstacle avoidance path can be an obstacle cost, a smoothness cost, or a lane cost.

[0071] In step 104, the plurality of obstacle avoidance paths are screened according to the driving costs respectively corresponding to the plurality of obstacle avoidance paths to obtain a target obstacle avoidance path.

[0072] The smaller the driving cost, the better the effect of the corresponding obstacle avoidance path. Therefore, the obstacle avoidance path with the smallest driving cost among multiple obstacle avoidance paths can be determined as the target obstacle avoidance path, so as to achieve the screening of multiple obstacle avoidance paths.

[0073] In some embodiments, after calculating the driving costs corresponding to multiple obstacle avoidance paths, the path planning method for autonomous driving also includes: when the target obstacle avoidance path cannot be screened out according to the driving costs corresponding to the multiple obstacle avoidance paths, discarding the farthest scattered waypoints, and returning to execute the step of planning multiple obstacle avoidance paths according to multiple discrete waypoints.

[0074] Here, it may happen that the target obstacle avoidance path cannot be screened out according to the driving costs corresponding to the multiple obstacle avoidance paths, for example, the driving costs corresponding to the multiple obstacle avoidance paths are the same, or the driving costs corresponding to the multiple obstacle avoidance paths are infinite. In this case, all the current farthest scattered road points are discarded, and the step of planning multiple obstacle avoidance paths according to multiple discrete road points is returned. For example, if the multiple discrete road points obtained by sampling processing are arranged in the form of M rows and N columns, the discrete road points located in the Mth row (the farthest scattered road points) are discarded, and the multiple obstacle avoidance paths are replanned according to the remaining M-1 rows and N columns of discrete road points. In this way, the discrete road points in the M-1th row are used as the new farthest scattered road points to ensure that the target obstacle avoidance path can be finally screened out, which greatly enhances the robustness of the path planning.

[0075] In step 105, the vehicle is controlled to travel along the target obstacle avoidance path.

[0076] Since the target obstacle avoidance path is the best among multiple obstacle avoidance paths, the vehicle is controlled to drive according to the target obstacle avoidance path.

[0077] In some embodiments, the above steps 101 to 105 may be performed periodically, that is, path planning is continuously performed during the driving of the vehicle to ensure that the latest target obstacle avoidance path can meet the latest situation and fully adapt to the ever-changing external environment.

[0078] like Figure 3A As shown, the embodiment of the present application establishes a solution space for obstacle avoidance paths based on the obstacle information in the lane where the vehicle is located and in the adjacent lanes; performs sampling processing in the solution space to obtain multiple discrete waypoints, and plans multiple obstacle avoidance paths based on the multiple discrete waypoints; wherein each obstacle avoidance path starts from the position of the vehicle and ends at the farthest scattered waypoint, and the farthest scattered waypoint represents the discrete waypoint with the largest longitudinal distance from the vehicle among the multiple discrete waypoints; calculates the driving costs corresponding to the multiple obstacle avoidance paths; screens the multiple obstacle avoidance paths according to the driving costs corresponding to the multiple obstacle avoidance paths to obtain the target obstacle avoidance path; and controls the vehicle to drive according to the target obstacle avoidance path. The embodiment of the present application comprehensively considers multiple lanes rather than being limited to the lane itself, making it convenient to make a decision on whether to avoid obstacles in the lane itself or to change lanes, and finally obtains the target obstacle avoidance path with the lowest driving cost, which can maximize driving efficiency while ensuring driving safety.

[0079] In some embodiments, see Figure 3B , Figure 3B is a flow chart of a path planning method for autonomous driving provided in an embodiment of the present application. Figure 3A The illustrated step 103 can be implemented through steps 201 to 204 , which will be described in conjunction with each step.

[0080] In step 201, an obstacle cost of any obstacle avoidance path is calculated according to lateral distances between a plurality of discrete waypoints in any obstacle avoidance path and obstacles.

[0081] For the sake of convenience, any obstacle avoidance path pathA among multiple obstacle avoidance paths is taken as an example to illustrate the process of calculating the driving cost.

[0082] On the one hand, the obstacle cost of the obstacle avoidance path pathA is calculated based on the lateral distances between the multiple discrete waypoints in the obstacle avoidance path pathA and the obstacles. The smaller the lateral distances between the multiple discrete waypoints in the obstacle avoidance path pathA and the obstacles, the greater the probability of collision with the obstacle, and therefore the greater the obstacle cost of the obstacle avoidance path pathA.

[0083] It is worth noting that the premise for calculating the lateral distance between a discrete waypoint and an obstacle is that the obstacle and the discrete waypoint are located in the same cross section (ie, in the same row).

[0084] In some embodiments, the above-mentioned calculation of the obstacle cost of any obstacle avoidance path based on the lateral distances between multiple discrete waypoints in any obstacle avoidance path and the obstacle can be achieved in the following way: the obstacle cost of any discrete waypoint in any obstacle avoidance path is calculated based on the lateral distance between any discrete waypoint in any obstacle avoidance path and the obstacle; a second fusion processing is performed on the obstacle costs corresponding to the multiple discrete waypoints in any obstacle avoidance path to obtain the obstacle cost of any obstacle avoidance path.

[0085] Here, the obstacle cost of each discrete waypoint in the obstacle avoidance path pathA is calculated based on the lateral distance between the discrete waypoint and the obstacle. Then, the obstacle costs corresponding to the multiple discrete waypoints in the obstacle avoidance path pathA are subjected to a second fusion process to obtain the obstacle cost of the obstacle avoidance path pathA. The fusion process (such as the first fusion process, the second fusion process, and the third fusion process) involved in the embodiments of the present application can be performed by summing, weighted summing, taking the maximum value, taking the average value, integrating, and the like. In the above method, each discrete waypoint in the obstacle avoidance path is considered separately, and then all the discrete waypoints in the obstacle avoidance path are combined to obtain the obstacle cost of the obstacle avoidance path, which provides a feasible calculation method for the obstacle cost of the obstacle avoidance path.

[0086] In some embodiments, the above-mentioned calculation of the obstacle cost of any discrete waypoint based on the lateral distance between any discrete waypoint and the obstacle in any obstacle avoidance path can be achieved in the following manner: when the lateral distance between any discrete waypoint and the obstacle is less than the minimum safety distance, the obstacle cost of any discrete waypoint is determined to be infinite; when the lateral distance between any discrete waypoint and the obstacle is greater than the maximum safety distance, the obstacle cost of any discrete waypoint is determined to be zero; when the lateral distance between any discrete waypoint and the obstacle is greater than or equal to the minimum safety distance and less than or equal to the maximum safety distance, the difference from the lateral distance between any discrete waypoint and the obstacle to the maximum safety distance is calculated, and the obstacle cost of any discrete waypoint is calculated based on the difference.

[0087] For the sake of convenience, take any discrete waypoint pointA in the obstacle avoidance path pathA as an example to illustrate the process of calculating the obstacle cost:

[0088] 1) When the lateral distance between the discrete waypoint pointA and the obstacle is less than the minimum safe distance, it is considered that there is a collision risk, so the obstacle cost of the discrete waypoint pointA is determined to be infinite. For example, it can be described by the following formula:

[0089] d_obs–vehicle_width / 2 <d_safe

[0090] Among them, d_obs represents the lateral distance between the center point of the vehicle and the obstacle when it is located at the discrete waypoint pointA; vehicle_width represents the width of the vehicle; d_obs-vehicle_width / 2 represents the lateral distance between the discrete waypoint pointA and the obstacle, that is, the redundant vehicle_width / 2 is removed on the basis of d_obs; d_safe represents the minimum safety distance, which can be set in advance.

[0091] 2) When the lateral distance between the discrete waypoint pointA and the obstacle is greater than the maximum safe distance, it is considered that there is no collision risk, so the obstacle cost of the discrete waypoint pointA is determined to be zero. For example, it can be described by the following formula:

[0092] d_obs–vehicle_width / 2>d_free

[0093] Among them, d_free represents the maximum safety distance and can be set in advance.

[0094] It is worth noting that the maximum safety distance is greater than the minimum safety distance.

[0095] 3) When the lateral distance between the discrete waypoint pointA and the obstacle is greater than or equal to the minimum safety distance and less than or equal to the maximum safety distance, the difference between the lateral distance between the discrete waypoint pointA and the obstacle and the maximum safety distance is calculated, and the obstacle cost of the discrete waypoint pointA is calculated based on the difference. The obstacle cost of the discrete waypoint pointA is positively correlated with the above difference. For example, it can be described by the following formula:

[0096] cost_obs=w_obs(d_free–d_obs)^2

[0097] Among them, cost_obs represents the obstacle cost of discrete waypoint pointA; w_obs represents the preset weight.

[0098] In some embodiments, the path planning method for autonomous driving also includes: determining a minimum safety distance range based on the obstacle type of the obstacle corresponding to any discrete waypoint; wherein the minimum value in the minimum safety distance range corresponds to the minimum obstacle avoidance speed of the vehicle, and the maximum value in the minimum safety distance range corresponds to the maximum obstacle avoidance speed of the vehicle; interpolation processing is performed based on the speed of the vehicle, the minimum obstacle avoidance speed of the vehicle, the maximum obstacle avoidance speed of the vehicle, and the minimum safety distance range to obtain the minimum safety distance.

[0099] The minimum safety distance may be preset or calculated by other methods. Here, an example is given for the latter method.

[0100] First, the minimum safety distance range is determined according to the obstacle type of the obstacle corresponding to the discrete waypoint pointA, wherein the obstacle corresponding to the discrete waypoint pointA refers to the obstacle located in the same cross section as the discrete waypoint pointA. For different obstacle types, the minimum safety distance range dist_min to dist_max can be pre-calibrated according to the maximum obstacle avoidance speed v_max and the minimum obstacle avoidance speed v_min of the vehicle, wherein the maximum obstacle avoidance speed refers to the maximum speed during the process of turning on the obstacle avoidance function, and the minimum obstacle avoidance speed refers to the minimum speed during the process of turning on the obstacle avoidance function, and the maximum obstacle avoidance speed and the minimum obstacle avoidance speed can be pre-set; the minimum value dist_min in the minimum safety distance range corresponds to v_min, and the maximum value dist_max in the minimum safety distance range corresponds to v_max. For example, obstacle types include static obstacles (such as water barriers, cones and cylinders) and dynamic obstacles (such as vehicles, two-wheelers and people). The minimum safe distance range for calibrated static obstacles is static_dist_min~static_dist_max, and the minimum safe distance range for calibrated dynamic obstacles is dynamic_dist_min~dynamic_dist_max.

[0101] After determining the minimum safety distance range dist_min to dist_max (for example, static_dist_min to static_dist_max or dynamic_dist_min to dynamic_dist_max, depending on the obstacle type), interpolation processing is performed based on the speed of the vehicle, the minimum obstacle avoidance speed of the vehicle, the maximum obstacle avoidance speed of the vehicle, and the minimum safety distance range. Through interpolation processing, a value corresponding to the speed of the vehicle is determined in the minimum safety distance range as the minimum safety distance. For example, it can be described by the following formula:

[0102] d_safe=(v–v_min)*(dist_max–dist_min) / (v_max–v_min)+dist_min

[0103] Wherein, v represents the speed of the vehicle (current speed).

[0104] In the above method, the appropriate minimum safety distance is determined according to the type of obstacle, which not only ensures driving safety and enhances the sense of security of the driver and passengers, but also ensures that the reasonable obstacle avoidance path will not be missed.

[0105] In step 202, the smoothness cost of any obstacle avoidance path is calculated according to the lateral distances between a plurality of discrete waypoints in any obstacle avoidance path.

[0106] On the other hand, the smoothness cost of the obstacle avoidance path pathA is calculated based on the lateral distances between multiple discrete waypoints in the obstacle avoidance path pathA, wherein, if the lateral distances between multiple discrete waypoints in the obstacle avoidance path pathA are smaller, it proves that the obstacle avoidance path pathA is smoother and the driving experience is more comfortable, so the smoothness cost of the obstacle avoidance path pathA is smaller.

[0107] In some embodiments, the above-mentioned calculation of the smoothness cost of any obstacle avoidance path based on the lateral distances between multiple discrete waypoints in any obstacle avoidance path can be achieved in the following manner: any obstacle avoidance path is split into multiple sub-paths; wherein each sub-path is composed of two adjacent discrete waypoints in any obstacle avoidance path; the smoothness cost of the sub-path is calculated based on the lateral distance between two discrete waypoints in each sub-path; when the smoothness cost of any sub-path is greater than a smoothness cost threshold, the smoothness cost of any obstacle avoidance path is determined to be infinite; when the smoothness costs corresponding to multiple sub-paths are all less than or equal to the smoothness cost threshold, the smoothness costs corresponding to the multiple sub-paths are subjected to a third fusion processing to obtain the smoothness cost of any obstacle avoidance path.

[0108] Here, the obstacle avoidance path pathA is first split into multiple sub-paths, where each sub-path is composed of two adjacent discrete waypoints in the obstacle avoidance path pathA. For example, if the discrete waypoints in the obstacle avoidance path pathA are pointA, pointB, pointC, and pointD, the obstacle avoidance path pathA can be split into three sub-paths, where sub-path pathA1 includes pointA and pointB, sub-path pathA2 includes pointB and pointC, and sub-path pathA3 includes pointC and pointD.

[0109] Then, according to the lateral distance between two discrete waypoints in each subpath, the smoothness cost of the subpath is calculated, wherein the smoothness cost of the subpath is positively correlated with the lateral distance between the two discrete waypoints in the subpath. For example, the lateral distance between the two discrete waypoints in the subpath may be divided by the longitudinal distance between the two discrete waypoints in the subpath to obtain the smoothness cost of the subpath.

[0110] When the smoothness cost of any sub-path is greater than the smoothness cost threshold, it proves that the obstacle avoidance path pathA is not smooth as a whole, and therefore the smoothness cost of the obstacle avoidance path pathA is determined to be infinite; when the smoothness costs corresponding to multiple sub-paths are less than or equal to the smoothness cost threshold, the smoothness costs corresponding to the multiple sub-paths are subjected to a third fusion process to obtain the smoothness cost of the obstacle avoidance path pathA.

[0111] For example, it can be described by the formula:

[0112] cost_smooth A1=(pointB_x–pointA_x) / delta_s

[0113] cost_smooth A2=(pointC_x–pointB_x) / delta_s

[0114] cost_smooth A3=(pointD_x–pointC_x) / delta_s

[0115] Among them, cost_smooth A1 represents the smoothness cost of subpath pathA1, pointA_x represents the lateral position of pointA, and so on; delta_s represents the longitudinal distance between two adjacent discrete waypoints in the obstacle avoidance path pathA (here it is assumed that the longitudinal distance between two discrete waypoints in each subpath is the same), which corresponds to the sampled longitudinal distance above.

[0116] When any one of cost_smooth A1, cost_smooth A2 and cost_smooth A3 is greater than the smoothness cost threshold, the smoothness cost of the obstacle avoidance path pathA is determined to be infinite; when cost_smooth A1, cost_smoothA2 and cost_smooth A3 are all less than or equal to the smoothness cost threshold, cost_smooth A1, cost_smoothA2 and cost_smooth A3 are subjected to a third fusion process to obtain the smoothness cost of the obstacle avoidance path pathA, that is, cost_smooth = w1*cost_smooth A1^2+w2*cost_smooth A2^2+w3*cost_smooth A3^2, wherein cost_smooth represents the smoothness cost of the obstacle avoidance path pathA, and w1, w2 and w3 represent pre-set weights.

[0117] In step 203, the lane cost of any obstacle avoidance path is calculated based on the lateral distances between the farthest scattered road points in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane.

[0118] On the other hand, the lane cost of the obstacle avoidance path pathA is calculated based on the lateral distances between the farthest scattered points in the obstacle avoidance path pathA and the lane where the vehicle is located and the adjacent lane. The larger the lateral distances between the farthest scattered points in the obstacle avoidance path pathA and the lane where the vehicle is located and the adjacent lane, the more difficult it is for the vehicle to return to the lane, and therefore the greater the lane cost of the obstacle avoidance path pathA.

[0119] In some embodiments, the lateral distance between the farthest scatter point and the lane may refer to the lateral distance between the farthest scatter point and the center line (longitudinal center line) of the lane. Based on this, the lane cost refers to the cost required for the vehicle to return to the center line of the lane.

[0120] In some embodiments, the above-mentioned calculation of the lane cost of any obstacle avoidance path based on the lateral distances between the farthest scattered point in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane can be achieved in the following way: based on the lateral distances between the farthest scattered point in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane, the nearest lane is selected from the lane where the vehicle is located and the adjacent lanes; based on the lateral distance between the farthest scattered point in any obstacle avoidance path and the nearest lane, the lane cost of any obstacle avoidance path is calculated.

[0121] For example, after determining the lateral distances between the farthest scattered point in the obstacle avoidance path pathA and the lane where the vehicle is located and the adjacent lane, the lane corresponding to the smallest lateral distance is determined as the nearest lane. Then, based on the lateral distance between the farthest scattered point in the obstacle avoidance path pathA and the nearest lane, the lane cost of the obstacle avoidance path pathA is calculated, for example:

[0122] cost_ref=w_ref*(l_cur-l_ref)^2

[0123] Among them, cost_ref represents the lane cost of the obstacle avoidance path pathA; w_ref represents the pre-set weight; l_cur represents the lateral position of the farthest scattered point in the obstacle avoidance path pathA; l_ref represents the lateral position of the center line of the nearest lane.

[0124] The above method selects the nearest lane to calculate the lane cost of the obstacle avoidance path under the premise that there are multiple lanes. This can conform to the actual vehicle driving rules, that is, the vehicle usually returns to the center line of the nearest lane to continue driving, thereby ensuring the accuracy of the obtained lane cost.

[0125] In step 204, a first fusion process is performed on the obstacle cost, the smoothness cost, and the lane centerline cost of any obstacle avoidance path to obtain the driving cost of any obstacle avoidance path.

[0126] Finally, the obstacle cost, smoothness cost, and lane centerline cost of the obstacle avoidance path pathA are first fused to obtain the driving cost of the obstacle avoidance path pathA, for example:

[0127] Cost=cost_obs+cost_smooth+cost_ref

[0128] Among them, Cost represents the driving cost of the obstacle avoidance path pathA; cost_obs in this formula represents the obstacle cost of the obstacle avoidance path pathA.

[0129] like Figure 3BAs shown, the embodiment of the present application calculates the obstacle cost of any obstacle avoidance path according to the lateral distances between multiple discrete waypoints in any obstacle avoidance path and the obstacle; calculates the smoothness cost of any obstacle avoidance path according to the lateral distances between multiple discrete waypoints in any obstacle avoidance path; calculates the lane cost of any obstacle avoidance path according to the lateral distances between the farthest discrete waypoints in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane; performs a first fusion process on the obstacle cost, smoothness cost and lane centerline cost of any obstacle avoidance path to obtain the driving cost of any obstacle avoidance path. In this way, the final driving cost is obtained by comprehensively considering the three costs, which can improve the comprehensiveness and accuracy of the driving cost, and facilitate the subsequent screening of the target obstacle avoidance path with the best overall effect according to the driving cost.

[0130] In some embodiments, see Figure 3C , Figure 3C is a flow chart of a path planning method for autonomous driving provided in an embodiment of the present application. Figure 3A The illustrated step 101 can be implemented through steps 301 to 302 , which will be described in conjunction with each step.

[0131] In step 301, the obstacle trajectory is predicted based on the obstacle information in the lane where the vehicle is located and the adjacent lanes.

[0132] Here, considering that obstacles may be in dynamic motion in traffic scenes, the obstacle trajectory is predicted based on the obstacle information in the lane where the vehicle is located and the adjacent lanes.

[0133] In some embodiments, the obstacle information includes the obstacle position and the obstacle speed. The above-mentioned prediction of the obstacle trajectory based on the obstacle information in the lane where the vehicle is located and the adjacent lane can be achieved in the following manner: when the obstacle speed of any obstacle in the lane where the vehicle is located and the adjacent lane is less than a preset speed, a uniform deceleration motion model is used to predict the obstacle position and obstacle speed of any obstacle to obtain the obstacle trajectory of any obstacle; when the obstacle speed of any obstacle is greater than or equal to the preset speed, a uniform speed motion model is used to predict the obstacle position and obstacle speed of any obstacle to obtain the obstacle trajectory of any obstacle.

[0134] Here, take obstacle A in the lane where the vehicle is located and any obstacle in the adjacent lane as an example. When the obstacle speed of obstacle A is less than the preset speed, it proves that it is most likely decelerating. Therefore, a uniform deceleration motion model (i.e., decelerating at a constant deceleration until the speed is zero) is used to predict the obstacle position and obstacle speed of obstacle A to obtain the obstacle trajectory of obstacle A; when the obstacle speed of obstacle A is greater than or equal to the preset speed, a uniform motion model is used to predict the obstacle position and obstacle speed of obstacle A to obtain the obstacle trajectory of obstacle A. In the above method, using a suitable motion model for prediction based on the obstacle speed of the obstacle can improve the accuracy of the predicted obstacle trajectory.

[0135] In some embodiments, the above-mentioned prediction of obstacle trajectory based on obstacle information in the lane where the vehicle is located and in adjacent lanes can be achieved in the following manner: the obstacle trajectory within a third preset time period is predicted based on obstacle information in the lane where the vehicle is located and in adjacent lanes. The third preset time period can be set according to the actual application scenario, such as being set to 6 to 8 seconds. On the one hand, considering that the size of the solution space to be established is limited, only predicting the obstacle trajectory within the third preset time period can save computing costs; on the other hand, the predicted obstacle trajectory within the third preset time period can ensure accuracy, while it is difficult to ensure accuracy if the obstacle trajectory that is farther in time is to be predicted.

[0136] In step 302, the space in the lane where the vehicle is located and the adjacent lanes and outside the obstacle track is determined as the solution space of the obstacle avoidance path.

[0137] Here, the space in the lane where the vehicle is located and the adjacent lanes and outside the obstacle trajectory (i.e., without obstacle trajectory) is determined as the solution space of the obstacle avoidance path, thereby ensuring that the planned obstacle avoidance path will not collide with obstacles in the future.

[0138] like Figure 3C As shown, the embodiment of the present application predicts the obstacle trajectory based on the obstacle information in the lane where the vehicle is located and the adjacent lanes; the space in the lane where the vehicle is located and the adjacent lanes and outside the obstacle trajectory is determined as the solution space of the obstacle avoidance path. In this way, considering the characteristics that the obstacle may be in dynamic motion, the obstacle trajectory is excluded when establishing the solution space, thereby ensuring that the obstacle avoidance path planned based on the solution space will not collide with the obstacle in the future.

[0139] The following is an exemplary structure of the automatic driving path planning device 455 provided in the embodiment of the present application implemented as a software module. In some embodiments, such as Figure 2 As shown, the software modules in the path planning device 455 of the automatic driving stored in the memory 450 may include: an establishment module 4551, which is used to establish a solution space for the obstacle avoidance path according to the obstacle information in the lane where the ego vehicle is located and the adjacent lanes; a planning module 4552, which is used to perform sampling processing in the solution space to obtain multiple discrete waypoints, and plan multiple obstacle avoidance paths according to the multiple discrete waypoints; wherein each obstacle avoidance path starts from the position of the ego vehicle and ends at the farthest scattered waypoint, and the farthest scattered waypoint represents the discrete waypoint with the largest longitudinal distance from the ego vehicle among the multiple discrete waypoints; a calculation module 4553, which is used to calculate the driving costs corresponding to the multiple obstacle avoidance paths; a screening module 4554, which is used to screen the multiple obstacle avoidance paths according to the driving costs corresponding to the multiple obstacle avoidance paths, and obtain a target obstacle avoidance path; a control module 4555, which is used to control the ego vehicle to travel according to the target obstacle avoidance path.

[0140] In some embodiments, the calculation module 4553 is also used to: perform the following processing for any obstacle avoidance path among the multiple obstacle avoidance paths: calculate the obstacle cost of any obstacle avoidance path based on the lateral distances between multiple discrete waypoints in any obstacle avoidance path and the obstacle; calculate the smoothness cost of any obstacle avoidance path based on the lateral distances between multiple discrete waypoints in any obstacle avoidance path; calculate the lane cost of any obstacle avoidance path based on the lateral distances between the farthest scattered waypoints in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane; perform a first fusion processing on the obstacle cost, smoothness cost and lane centerline cost of any obstacle avoidance path to obtain the driving cost of any obstacle avoidance path.

[0141] In some embodiments, the calculation module 4553 is also used to: calculate the obstacle cost of any discrete waypoint in any obstacle avoidance path based on the lateral distance between any discrete waypoint and the obstacle; perform a second fusion processing on the obstacle costs corresponding to multiple discrete waypoints in any obstacle avoidance path to obtain the obstacle cost of any obstacle avoidance path.

[0142] In some embodiments, the calculation module 4553 is also used to: when the lateral distance between any discrete waypoint and the obstacle is less than the minimum safety distance, determine that the obstacle cost of any discrete waypoint is infinite; when the lateral distance between any discrete waypoint and the obstacle is greater than the maximum safety distance, determine that the obstacle cost of any discrete waypoint is zero; when the lateral distance between any discrete waypoint and the obstacle is greater than or equal to the minimum safety distance and less than or equal to the maximum safety distance, calculate the difference from the lateral distance between any discrete waypoint and the obstacle to the maximum safety distance, and calculate the obstacle cost of any discrete waypoint based on the difference.

[0143] In some embodiments, the calculation module 4553 is also used to: determine the minimum safety distance range according to the obstacle type of the obstacle corresponding to any discrete waypoint; wherein the minimum value in the minimum safety distance range corresponds to the minimum obstacle avoidance speed of the vehicle, and the maximum value in the minimum safety distance range corresponds to the maximum obstacle avoidance speed of the vehicle; interpolation processing is performed according to the speed of the vehicle, the minimum obstacle avoidance speed of the vehicle, the maximum obstacle avoidance speed of the vehicle and the minimum safety distance range to obtain the minimum safety distance.

[0144] In some embodiments, the calculation module 4553 is also used to: split any obstacle avoidance path into multiple sub-paths; wherein each sub-path is composed of two adjacent discrete waypoints in any obstacle avoidance path; calculate the smoothness cost of the sub-path based on the lateral distance between the two discrete waypoints in each sub-path; when the smoothness cost of any sub-path is greater than the smoothness cost threshold, determine that the smoothness cost of any obstacle avoidance path is infinite; when the smoothness costs corresponding to multiple sub-paths are all less than or equal to the smoothness cost threshold, perform a third fusion processing on the smoothness costs corresponding to the multiple sub-paths to obtain the smoothness cost of any obstacle avoidance path.

[0145] In some embodiments, the calculation module 4553 is also used to: select the nearest lane from the lane where the vehicle is located and the adjacent lanes based on the lateral distances between the farthest scattered road points in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lanes; calculate the lane cost of any obstacle avoidance path based on the lateral distance between the farthest scattered road points in any obstacle avoidance path and the nearest lane.

[0146] In some embodiments, module 4551 is established and is also used to: predict the obstacle trajectory based on the obstacle information in the lane where the vehicle is located and the adjacent lanes; and determine the space in the lane where the vehicle is located and the adjacent lanes and outside the obstacle trajectory as the solution space for the obstacle avoidance path.

[0147] In some embodiments, the obstacle information includes the obstacle position and the obstacle speed; the module 4551 is established, and is also used for: when the obstacle speed of any obstacle in the lane where the vehicle is located and the adjacent lane is less than a preset speed, a uniform deceleration motion model is used to predict the obstacle position and obstacle speed of any obstacle to obtain the obstacle trajectory of any obstacle; when the obstacle speed of any obstacle is greater than or equal to the preset speed, a uniform speed motion model is used to predict the obstacle position and obstacle speed of any obstacle to obtain the obstacle trajectory of any obstacle.

[0148] In some embodiments, the path planning device 455 for autonomous driving also includes an acquisition module, which is used to: determine a first longitudinal distance based on the speed of the vehicle and a first preset time; and obtain obstacle information of obstacles in the lane where the vehicle is located and in adjacent lanes, and whose longitudinal distances to the vehicle are less than the first longitudinal distance.

[0149] In some embodiments, the planning module 4552 is further used to: perform sampling processing in the longitudinal direction of the solution space according to the sampled longitudinal distance, and in the transverse direction of the solution space according to the sampled transverse distance, to obtain multiple discrete waypoints.

[0150] In some embodiments, the planning module 4552 is also used to: determine the second longitudinal distance based on the speed of the vehicle and the second preset time; and determine the maximum value of the second longitudinal distance and the preset longitudinal distance as the sampling longitudinal distance for sampling processing in the longitudinal direction.

[0151] In some embodiments, the path planning device 455 for autonomous driving also includes a discarding module, which is used to: when the target obstacle avoidance path cannot be screened out according to the driving costs corresponding to multiple obstacle avoidance paths, discard the farthest scattered waypoint and return to execute the step of planning multiple obstacle avoidance paths according to multiple discrete waypoints.

[0152] The embodiment of the present application provides a computer program product or a computer program, which includes executable instructions stored in a computer-readable storage medium. The processor of the electronic device reads the executable instructions from the computer-readable storage medium, and the processor executes the executable instructions, so that the electronic device implements the above-mentioned path planning method for autonomous driving in the embodiment of the present application.

[0153] An embodiment of the present application provides a computer-readable storage medium storing executable instructions, wherein executable instructions are stored. When the executable instructions are executed by a processor, the processor will implement the path planning method for autonomous driving provided by the embodiment of the present application.

[0154] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface storage, optical disk, or CD-ROM; or it may be various devices including one or any combination of the above memories.

[0155] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0156] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0157] The above are only embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. A path planning method for autonomous driving, characterized in that: include: Establish a solution space for obstacle avoidance paths based on the obstacle information in the lane where the vehicle is located and in adjacent lanes; Sampling is performed in the solution space to obtain a plurality of discrete waypoints, and a plurality of obstacle avoidance paths are planned according to the plurality of discrete waypoints; wherein each obstacle avoidance path starts from the position of the ego vehicle and ends at the most distant discrete waypoint, and the most distant discrete waypoint represents a discrete waypoint with the largest longitudinal distance from the ego vehicle among the plurality of discrete waypoints; Calculating the driving costs corresponding to the plurality of obstacle avoidance paths respectively; Filtering the plurality of obstacle avoidance paths according to the driving costs respectively corresponding to the plurality of obstacle avoidance paths to obtain a target obstacle avoidance path; The vehicle is controlled to travel according to the target obstacle avoidance path.

2. The method according to claim 1, characterized in that: The calculating the driving costs respectively corresponding to the plurality of obstacle avoidance paths includes: For any obstacle avoidance path among the multiple obstacle avoidance paths, perform the following processing: Calculating an obstacle cost of any obstacle avoidance path according to lateral distances between a plurality of discrete waypoints in any obstacle avoidance path and obstacles; Calculating a smoothness cost of any obstacle avoidance path according to lateral distances between a plurality of discrete waypoints in any obstacle avoidance path; Calculating a lane cost of any obstacle avoidance path according to the lateral distances between the farthest scattered road points in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane; A first fusion process is performed on the obstacle cost, the smoothness cost, and the lane centerline cost of any obstacle avoidance path to obtain the driving cost of any obstacle avoidance path.

3. The method according to claim 2, characterized in that The calculating the obstacle cost of any obstacle avoidance path according to the lateral distances between the plurality of discrete waypoints in any obstacle avoidance path and the obstacles includes: Calculating an obstacle cost of any discrete waypoint in any obstacle avoidance path according to a lateral distance between the discrete waypoint and the obstacle; A second fusion process is performed on obstacle costs respectively corresponding to a plurality of discrete waypoints in any obstacle avoidance path to obtain the obstacle cost of any obstacle avoidance path.

4. The method according to claim 3, characterized in that The calculating the obstacle cost of any discrete waypoint in any obstacle avoidance path according to the lateral distance between the discrete waypoint and the obstacle includes: When the lateral distance between any discrete waypoint and the obstacle is less than the minimum safety distance, determining the obstacle cost of any discrete waypoint to be infinite; When the lateral distance between any discrete waypoint and the obstacle is greater than the maximum safety distance, determining the obstacle cost of any discrete waypoint to be zero; When the lateral distance between any discrete waypoint and the obstacle is greater than or equal to the minimum safety distance and less than or equal to the maximum safety distance, the difference between the lateral distance between any discrete waypoint and the obstacle and the maximum safety distance is calculated, and the obstacle cost of any discrete waypoint is calculated based on the difference.

5. The method according to claim 4, characterized in that The method further comprises: Determine a minimum safety distance range according to the obstacle type of the obstacle corresponding to any discrete waypoint; wherein the minimum value in the minimum safety distance range corresponds to the minimum obstacle avoidance speed of the vehicle, and the maximum value in the minimum safety distance range corresponds to the maximum obstacle avoidance speed of the vehicle; The minimum safety distance is obtained by performing interpolation processing according to the speed of the vehicle, the minimum obstacle avoidance speed of the vehicle, the maximum obstacle avoidance speed of the vehicle, and the minimum safety distance range.

6. The method according to claim 2, characterized in that The calculating the smoothness cost of any obstacle avoidance path according to the lateral distances between the plurality of discrete waypoints in any obstacle avoidance path comprises: Splitting any obstacle avoidance path into multiple sub-paths; wherein each sub-path is composed of two adjacent discrete waypoints in any obstacle avoidance path; Calculate the smoothness cost of each subpath according to the lateral distance between two discrete waypoints in the subpath; When the smoothness cost of any sub-path is greater than the smoothness cost threshold, determining that the smoothness cost of any obstacle avoidance path is infinite; When the smoothness costs respectively corresponding to the multiple sub-paths are all less than or equal to the smoothness cost threshold, a third fusion process is performed on the smoothness costs respectively corresponding to the multiple sub-paths to obtain the smoothness cost of any obstacle avoidance path.

7. The method according to claim 2, characterized in that: The calculating the lane cost of any obstacle avoidance path according to the lateral distances between the farthest scattered road points in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane respectively includes: According to the lateral distances between the farthest scattered road points in any obstacle avoidance path and the lane where the vehicle is located and the adjacent lane, the nearest lane is selected from the lane where the vehicle is located and the adjacent lane; The lane cost of any obstacle avoidance path is calculated according to the lateral distance between the farthest scattered road point in any obstacle avoidance path and the nearest lane.

8. The method according to claim 1, characterized in that The solution space for establishing the obstacle avoidance path according to the obstacle information in the lane where the vehicle is located and the adjacent lanes includes: Obtaining an obstacle trajectory based on information about obstacles in the lane where the vehicle is located and in adjacent lanes; The space in the lane where the vehicle is located and the adjacent lanes and outside the obstacle trajectory is determined as the solution space of the obstacle avoidance path.

9. The method according to claim 8, characterized in that The obstacle information includes the obstacle position and obstacle speed; The step of predicting the obstacle trajectory according to the obstacle information in the lane where the vehicle is located and in the adjacent lanes includes: When the obstacle speed of any obstacle in the lane where the vehicle is located and the adjacent lane is less than the preset speed, a uniform deceleration motion model is used to predict the obstacle position and obstacle speed of any obstacle to obtain the obstacle trajectory of any obstacle; When the obstacle speed of any obstacle is greater than or equal to the preset speed, a uniform motion model is used to perform prediction according to the obstacle position and obstacle speed of any obstacle to obtain an obstacle trajectory of any obstacle.

10. The method according to claim 1, characterized in that Before establishing a solution space for an obstacle avoidance path according to obstacle information in the lane where the vehicle is located and in adjacent lanes, the method further includes: Determining a first longitudinal distance according to the speed of the vehicle and a first preset time duration; Obstacle information of obstacles in the lane where the vehicle is located and in adjacent lanes, the longitudinal distance between which and the vehicle is less than the first longitudinal distance, is obtained.

11. The method according to claim 1, characterized in that The sampling process is performed in the solution space to obtain a plurality of discrete waypoints, including: Sampling is performed in a longitudinal direction within the solution space according to a sampled longitudinal distance and in a transverse direction within the solution space according to a sampled transverse distance to obtain a plurality of discrete waypoints.

12. The method according to claim 11, characterized in that Before performing sampling processing in the longitudinal direction of the solution space according to the sampling longitudinal distance and in the transverse direction of the solution space according to the sampling transverse distance, the method further includes: Determining a second longitudinal distance according to the speed of the vehicle and a second preset time duration; The maximum value of the second longitudinal distance and the preset longitudinal distance is determined as the sampling longitudinal distance for performing sampling processing in the longitudinal direction.

13. The method according to claim 1, characterized in that The method further comprises: When the target obstacle avoidance path cannot be screened out according to the driving costs corresponding to the multiple obstacle avoidance paths, the farthest scattered waypoint is discarded, and the process returns to the step of planning multiple obstacle avoidance paths according to the multiple discrete waypoints.

14. A path planning device for automatic driving, characterized in that: include: Establish a module for establishing a solution space for an obstacle avoidance path based on information about obstacles in the lane where the vehicle is located and in adjacent lanes; A planning module, configured to perform sampling processing in the solution space to obtain a plurality of discrete waypoints, and plan a plurality of obstacle avoidance paths according to the plurality of discrete waypoints; wherein each obstacle avoidance path starts from the position of the ego vehicle and ends at the most distant discrete waypoint, wherein the most distant discrete waypoint represents a discrete waypoint with the largest longitudinal distance from the ego vehicle among the plurality of discrete waypoints; A calculation module, used to calculate the driving costs corresponding to the plurality of obstacle avoidance paths respectively; A screening module, used for screening the plurality of obstacle avoidance paths according to the driving costs respectively corresponding to the plurality of obstacle avoidance paths, to obtain a target obstacle avoidance path; The control module is used to control the vehicle to travel according to the target obstacle avoidance path.

15. An electronic device, characterized in that: include: A memory for storing executable instructions; A processor, configured to implement the method according to any one of claims 1 to 13 when executing the executable instructions stored in the memory.

16. A computer-readable storage medium, characterized in that: Executable instructions are stored, and when executed by a processor, the method described in any one of claims 1 to 13 is implemented.

Images