Unmanned vehicle speed planning control method and device, electronic equipment and storage medium

Abstract

The application provides a speed planning control method and device for unmanned vehicles, electronic equipment and storage medium. The method is applied to unmanned vehicles, unmanned driving equipment or automatic driving equipment, and includes: dividing a detection area around the unmanned vehicle according to driving state information and road environment information during driving of the unmanned vehicle; obtaining identification information of obstacles around the unmanned vehicle, and classifying the obstacles around the unmanned vehicle by using a predetermined classification method; determining whether there are obstacles in the detection area, and when there are obstacles in the detection area and the category of the obstacles in the detection area is a risky obstacle category, performing speed planning control on the unmanned vehicle by using a first speed planning method according to the relative speed and relative distance between the unmanned vehicle and the obstacles in the detection area; otherwise, performing speed planning control on the unmanned vehicle by using a second speed planning method. The application improves the success rate of speed planning and the smoothness and safety of vehicle operation.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to an unmanned vehicle speed planning and control method, device, electronic equipment and storage medium. Background Technology

[0002] Velocity planning control, based on local path planning and given one or more path curves, adds velocity-related information to meet the operational constraints of feedback control and conform to the output results of behavioral decisions. Its primary consideration is the avoidance of dynamic obstacles. Existing velocity planning methods include: generating velocity by specifying linear acceleration, spline interpolation, function fitting, target time point methods, and dynamic programming algorithms, among which dynamic programming is the most commonly used.

[0003] The input to dynamic programming algorithms includes environmental information such as the reference path and obstacles. Its computation process mainly includes obstacle prediction and processing, ST graph generation, ST graph sampling and search, and speed smoothing. However, current speed planning and control schemes still have some problems. While dynamic programming algorithms can successfully plan speed, they perform well on structured roads with only motor vehicles. This is because the safe distance between motor vehicles is relatively large, and their driving behavior is relatively standardized. However, in mixed traffic flow scenarios involving motor vehicles, bicycles, and pedestrians, there are diverse obstacle types, smaller safe distances to other vehicles, and irregular behaviors such as close overtaking and following other vehicles. In these complex situations, existing obstacle prediction and processing methods can easily lead to speed planning and control failures, potentially resulting in a series of comfort and safety issues such as sudden braking, rear-end collisions, and other accidents. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method, apparatus, electronic device and storage medium for speed planning and control of unmanned vehicles, in order to solve the problems that existing technologies are prone to speed planning and control failure in mixed traffic flow scenarios, which in turn cause a series of comfort and safety issues.

[0005] A first aspect of this application provides a speed planning and control method for an unmanned vehicle. The method includes: collecting driving state information and road environment information during the unmanned vehicle's operation; defining a detection area around the unmanned vehicle based on the driving state information and road environment information; classifying the obstacles around the unmanned vehicle using a predetermined classification method based on the identification information of obstacles around the unmanned vehicle by a perception module to determine the category of the obstacles; determining whether there are obstacles within the detection area; if there are obstacles within the detection area, and the category of the obstacles within the detection area is a risky obstacle category, performing speed planning and control on the unmanned vehicle using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and the obstacles within the detection area; otherwise, performing speed planning and control on the unmanned vehicle using a predetermined second speed planning method.

[0006] A second aspect of this application provides an unmanned vehicle speed planning and control device, comprising: a division module configured to collect driving state information and road environment information during the driving process of the unmanned vehicle, and to divide a detection area around the unmanned vehicle based on the driving state information and road environment information; a classification module configured to classify the obstacles around the unmanned vehicle using a predetermined classification method based on the identification information of obstacles around the unmanned vehicle by the perception module, so as to determine the category of the obstacles; and a speed planning and control module configured to determine whether there are obstacles in the detection area, and when there are obstacles in the detection area, and the category of the obstacles in the detection area is a risky obstacle category, to perform speed planning and control of the unmanned vehicle using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and the obstacles in the detection area; otherwise, to perform speed planning and control of the unmanned vehicle using a predetermined second speed planning method.

[0007] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0008] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0009] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects:

[0010] By collecting driving status information and road environment information during the autonomous vehicle's operation, a detection zone is delineated around the autonomous vehicle based on this information. The system then classifies the obstacles around the autonomous vehicle using a predetermined classification method based on the obstacle identification information from the perception module. It determines whether an obstacle exists within the detection zone. If an obstacle exists and is classified as a risky obstacle, the system performs speed planning and control of the autonomous vehicle using a predetermined first speed planning method based on the relative speed and distance between the autonomous vehicle and the obstacle. Otherwise, a predetermined second speed planning method is used for speed planning and control. This application improves the success rate of speed planning and enhances the smoothness and safety of the vehicle's operation. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of obstacle prediction and handling in existing technologies;

[0013] Figure 2 This is the ST diagram of motor vehicles traveling normally in the same direction in the existing technology scenario;

[0014] Figure 3 This is a schematic diagram of the prediction and processing results of obstacles in existing technologies;

[0015] Figure 4 This is an ST diagram generated when obstacles are relatively close to the autonomous vehicle in existing technology scenarios;

[0016] Figure 5 This is a flowchart illustrating the unmanned vehicle speed planning and control method provided in an embodiment of this application;

[0017] Figure 6 This is a schematic diagram illustrating the detection area defined around the unmanned vehicle in an embodiment of this application;

[0018] Figure 7 This is a schematic diagram illustrating the possible positional relationship between the obstacle and the collision detection area in front;

[0019] Figure 8 This is a schematic diagram illustrating the possible positional relationship between the obstacle and the side collision detection area;

[0020] Figure 9This is a schematic diagram of the structure of the unmanned vehicle speed planning and control device provided in the embodiments of this application;

[0021] Figure 10 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0022] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0023] Autonomous vehicles, also known as self-driving vehicles, driverless cars, or wheeled mobile robots, are integrated and intelligent new-era technological products that combine environmental perception, path planning, state recognition, and vehicle control. An autonomous driving system for an autonomous vehicle mainly consists of sensor-driven systems, a perception system, a fusion system, a decision-making system, a motion planning system, and a control system. In autonomous driving, the main role of speed planning and control is to ensure that the vehicle generates a speed adapted to the current driving environment based on the current environmental perception and vehicle state, while meeting constraints such as driving comfort, safety, and efficiency.

[0024] Velocity planning control, after providing one or more selected path curves for local path planning, requires incorporating velocity-related information into these local paths to meet the operational constraints of feedback control and conform to the output results of behavioral decisions. Essentially, it primarily considers the avoidance of dynamic obstacles. Existing velocity planning methods include: generating velocity by specifying linear acceleration, spline interpolation, function fitting, target time point methods, and dynamic programming algorithms. Among these, dynamic programming algorithms are the most commonly used.

[0025] The dynamic programming algorithm requires environmental information such as reference path and obstacles as input. The calculation process of the dynamic programming algorithm mainly includes obstacle prediction and processing, ST graph generation, ST graph sampling and search, and speed smoothing.

[0026] Obstacle prediction and handling refers to the situation where a moving obstacle appears in the environment and may occupy the planned path at some point in the future. If the target vehicle continues to travel at its current constant speed, it may collide with the moving obstacle at some point in the future. Therefore, it is necessary to calculate the range of t and s when a collision between the obstacle and the autonomous vehicle's planned trajectory is possible.

[0027] ST graph generation utilizes the aforementioned obstacle prediction information to mark a segment of path S occupied by an obstacle at a certain time Δt on the ST graph.

[0028] ST graph sampling and search first discretizes the ST graph, and then for each ST candidate target point, searches the ST search space to generate a path from the starting point to the target point.

[0029] Speed ​​smoothing is the smoothing of the tortuous and discrete path obtained in the ST graph sampling and search above, which satisfies the avoidance of dynamic obstacles while also meeting the requirements of smoothness, comfort and control of autonomous vehicles.

[0030] Based on the above introduction, it can be seen that obstacle prediction and autonomous vehicle trajectory planning are the data foundation required for the entire speed planning calculation. The success of speed planning using dynamic programming algorithms largely depends on the obstacle input. On structured roads with only motor vehicles, obstacles are other motor vehicles. Due to the relatively large safe distance between motor vehicles and their relatively standardized driving behavior, the speed planning method based on dynamic programming algorithms has a high success rate. However, in mixed traffic flow scenarios involving motor vehicles, bicycles, and pedestrians, there are more types of obstacles, smaller safe distances from the autonomous vehicle, and obstacles may exhibit irregular behaviors such as close overtaking and following. In such cases, using the obstacle prediction method described above can easily lead to speed planning failure, which can further cause a series of problems affecting vehicle comfort and safety, such as sudden braking, rear-end collisions, and other collisions.

[0031] The following section, with reference to the accompanying drawings, explains the specific implementation process of the existing speed planning method based on dynamic programming algorithms. For example... Figures 1 to 4 As shown, Figure 1 This is a schematic diagram of obstacle prediction and handling in existing technologies. Figure 2 This is the ST diagram of motor vehicles traveling normally in the same direction in the existing technology scenario. Figure 3 This is a schematic diagram of the prediction and processing results of obstacles in existing technology. Figure 4 This is an ST diagram generated when obstacles are relatively close to the autonomous vehicle in existing technology scenarios.

[0032] like Figure 1 This demonstrates the existing process for obstacle prediction. It calculates the time-path graph (ST graph) of the obstacle objects (OBs) by considering the predicted trajectories of the obstacles and the planned path of the autonomous vehicle's ego. For a vehicle traveling in the same direction and moving normally, its ST graph is as follows: Figure 2As shown, existing technologies can handle this using conventional speed planning methods. However, in real-world mixed traffic flow scenarios, especially when driving on auxiliary lanes, obstacles in the same direction may include not only motor vehicles but also non-motorized vehicles such as electric bicycles. When the autonomous vehicle's speed is relatively slow, the prediction and handling results for these obstacles are as follows... Figure 3 As shown. At this time, the distance between the obstacle (obs) and the autonomous vehicle's ego is relatively close, and the resulting ST diagram is as follows. Figure 4 As shown. From Figure 4 As can be seen, the solution space for velocity planning is extremely small or even non-existent. If dynamic programming is continued to be used for velocity planning at this time, it may lead to planning failure.

[0033] Based on the aforementioned existing technical solutions, it can be seen that in mixed traffic flow scenarios, relying solely on dynamic programming for speed planning cannot guarantee a high success rate and may even lead to a series of problems such as sudden braking, rear-end collisions, and other incidents. Therefore, while the existing dynamic programming algorithms have a high success rate for speed planning of vehicles traveling in the same direction, their success rate is low when non-motorized vehicles in the same direction exhibit irregular behavior, easily leading to a series of comfort and safety issues such as sudden braking, rear-end collisions, and other incidents. This application, in the obstacle prediction and processing process, divides a certain detection area around the vehicle body according to the autonomous vehicle's driving state. Combining the obstacle type and whether the obstacle is within the detection area, different speed planning control methods are selected for obstacles meeting different conditions, thereby ensuring a high success rate for speed planning. This allows for the stable handling of different types of obstacles while also improving the vehicle's comfort and safety.

[0034] The following will describe in detail an unmanned vehicle speed planning and control method and apparatus according to an embodiment of this application, with reference to the accompanying drawings and specific embodiments.

[0035] Figure 5 This is a flowchart illustrating the unmanned vehicle speed planning and control method provided in the embodiments of this application. Figure 5 The speed planning and control method for autonomous vehicles can be executed by the control unit of the autonomous vehicle's infotainment system. For example... Figure 5 As shown, the autonomous vehicle speed planning and control method may specifically include:

[0036] S501 collects driving status information and road environment information during the driving process of the unmanned vehicle, and delineates the detection area around the unmanned vehicle based on the driving status information and road environment information.

[0037] S502, Based on the identification information of obstacles around the unmanned vehicle by the perception module, the obstacles around the unmanned vehicle are classified using a predetermined classification method to determine the category of the obstacles;

[0038] S503, determine whether there is an obstacle in the detection area. If there is an obstacle in the detection area and the obstacle in the detection area is classified as a risky obstacle, perform speed planning control on the unmanned vehicle using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and the obstacle in the detection area; otherwise, perform speed planning control on the unmanned vehicle using a predetermined second speed planning method.

[0039] In some embodiments, a detection area is defined around the autonomous vehicle based on driving status information and road environment information, including: defining a frontal collision detection area and a side collision detection area in front of and on both sides of the autonomous vehicle based on driving status information and road environment information, and superimposing the frontal collision detection area and the side collision detection area to form a complete detection area.

[0040] Specifically, before defining the detection zone around the autonomous vehicle, various sensors installed on the autonomous vehicle are used to collect real-time information on the vehicle's driving status and the external road environment. For example, in practical applications, vehicle speed sensors and vision sensors can be used to obtain the autonomous vehicle's driving status information, which includes, but is not limited to, vehicle speed information, turning information, etc. In addition, data returned by various sensors (such as LiDAR) can be used to obtain real-time road environment information, which includes, but is not limited to, the width of the road ahead, whether the road is curved, whether the road ahead is straight or curved, etc.

[0041] After acquiring real-time driving status and road environment information of the autonomous vehicle, a detection zone is dynamically delineated around the vehicle based on its driving status and changes in the surrounding environment. The process and principle of this detection zone delineation are explained in detail below with reference to the accompanying diagram. Figure 6 This is a schematic diagram illustrating the division of a detection area around the unmanned vehicle according to an embodiment of this application. Figure 7 This is a schematic diagram illustrating the possible positional relationship between the obstacle and the collision detection area ahead. Figure 8 This is a schematic diagram illustrating the possible positional relationship between the obstacle and the side collision detection area. For example... Figures 6 to 8 As shown, the process of dividing the detection area can specifically include:

[0042] In this embodiment, a detection area can be defined around the autonomous vehicle based on its driving status information and surrounding road environment information. Specifically, a forward collision detection area and a side collision detection area can be defined in front of and to the sides of the autonomous vehicle based on this information, and then these two areas can be superimposed to form a complete detection area.

[0043] In one example, such as Figure 6 As shown, based on the driving status information of the autonomous vehicle, a specific detection area check_area is delineated around it. Figure 6 (A grayscale display area). The size parameters of this detection area are related to the size of the autonomous vehicle, its speed, and the road environment. This detection area can be formed by superimposing three rectangular areas: front, left, and right. Although a rectangular area is used in this embodiment, the shape of the area is not limited to a rectangle in practical applications.

[0044] Taking a rectangular area as an example, a frontal collision detection area (check_area_front) is defined in front of the autonomous vehicle, and side collision detection areas (check_area_side) are defined on the left and right sides of the vehicle. Obstacles can exist anywhere within the detection areas check_area_front and check_area_side. For example, the possible positional relationship between obstacles (obs) and the detection areas can be referenced. Figure 7 and Figure 8 As shown.

[0045] In some embodiments, the method further includes: the shape and size of the forward collision detection area and the side collision detection area change with the changes in the driving status information and road environment information of the autonomous vehicle, wherein the default shape of the forward collision detection area and the side collision detection area is a rectangle.

[0046] Specifically, in this embodiment, detection areas are defined in front of and to the sides of the vehicle based on the vehicle's driving status information and the surrounding road environment information. The shape and size of these detection areas can vary depending on the autonomous vehicle's driving status (e.g., speed) and road environment (e.g., whether there is an intersection or curve ahead). For example, if the vehicle is traveling at a high speed or there is an intersection ahead, the detection area in front may need to be expanded. Conversely, if the vehicle is traveling on a straight road at a slow speed, the detection area in front can be relatively small.

[0047] In this embodiment, the detection area is divided into a front collision detection area (check_area_front) and a side collision detection area (check_area_side). Finally, the rectangular areas corresponding to the front collision detection area and the side collision detection area (e.g., ...) are used to define the detection area. Figure 6 The three gray rectangular areas shown are superimposed to form a complete detection area. In practical applications, the shape and size of these three rectangular areas can be dynamically adjusted according to the size of the autonomous vehicle, its speed, and the road environment.

[0048] In some embodiments, based on the identification information of obstacles around the autonomous vehicle by the perception module, the obstacles around the autonomous vehicle are classified using a predetermined classification method, including: acquiring identification information obtained by the perception module of the autonomous vehicle from detecting and identifying obstacles around the autonomous vehicle, wherein the identification information includes obstacle type and obstacle size; classifying the obstacles around the autonomous vehicle according to the obstacle type and obstacle size, wherein the obstacle categories include a risk-free obstacle category and a risky obstacle category.

[0049] Specifically, in this embodiment, the obstacles around the autonomous vehicle are classified using a preset classification method based on the identification information obtained by the perception module. First, the identification information obtained by the perception module from detecting and identifying obstacles around the autonomous vehicle is acquired. This identification information includes at least the type information and size (e.g., cross-sectional area) of the obstacles. Then, the obstacles around the autonomous vehicle are classified according to their type and size. In this process, obstacles are mainly divided into two categories: non-risk obstacles and risky obstacles.

[0050] In one example, during the specific classification process, the participants in the mixed traffic flow are very complex, including trucks, buses, passenger cars, tricycles, motorcycles, electric vehicles, pedestrians, and balance scooters, etc. When the autonomous vehicle's speed is around 20 km / h, these types of obstacles may overtake the autonomous vehicle, thus affecting its speed planning. Therefore, when performing speed planning, it is particularly important to combine the results of the perception module to reasonably classify obstacles.

[0051] In the embodiments of this application, obstacles are classified into two categories based on whether they may cause irregular traffic behavior: obstacles with no potential risk (obs_safe, i.e., risk-free obstacle category) and obstacles with potential risk (obs_potential_risk, i.e., risky obstacle category). This classification method can more accurately assess the impact of obstacles on the speed planning of autonomous vehicles, thereby improving the driving safety and efficiency of autonomous vehicles in mixed traffic flows.

[0052] In some embodiments, obstacles around the autonomous vehicle are classified according to the type and size of the obstacle, including: determining whether the obstacle belongs to a motor vehicle or a non-motor vehicle based on the type of obstacle; when the obstacle is determined to belong to a non-motor vehicle, or the size of the obstacle is smaller than a preset condition, the obstacle is classified as a risky obstacle; otherwise, the obstacle is classified as a non-risky obstacle.

[0053] Specifically, this application embodiment classifies obstacles around the autonomous vehicle based on their type and size (e.g., cross-sectional area). In the obstacle type classification process, it first determines whether the obstacle belongs to a motor vehicle or a non-motor vehicle. Then, if the obstacle is determined to be a non-motor vehicle, or if the obstacle's cross-sectional area is smaller than a preset condition (e.g., a preset cross-sectional area threshold), the obstacle is classified as a risky obstacle, i.e., obs_potential_risk; otherwise, it is classified as a risk-free obstacle, i.e., obs_safe.

[0054] Furthermore, in the specific judgment process, the judgment criteria for obs_potential_risk are based on the type and cross-sectional area of ​​the obstacle. However, the classification criteria for obstacle types in this application are not limited to obstacle type and cross-sectional area. Any method that can distinguish obstacle types is applicable to the technical solution of this application, and this application does not limit it.

[0055] In one example, obstacles that are classified as non-motorized vehicles, or obstacles with a cross-sectional area less than 4 square meters, can be classified as obs_potential_risk (i.e., risky obstacles). Conversely, obstacles that are classified as motorized vehicles, or obstacles with a cross-sectional area greater than 4 square meters, can be classified as obs_safe (i.e., risk-free obstacles).

[0056] This application embodiment further selects appropriate speed planning methods for obstacles that meet different preset conditions based on the presence and type of obstacles in the detection area, thereby achieving speed planning control for different obstacle types within the detection area. The obstacle classification method of this application embodiment can more accurately evaluate obstacles in mixed traffic flows, thus contributing to more accurate speed planning and improving the driving safety and efficiency of autonomous vehicles in complex traffic environments.

[0057] In some embodiments, speed planning control of the autonomous vehicle is performed using a predetermined first speed planning method based on the relative speed and relative distance between the autonomous vehicle and obstacles within the detection area, including:

[0058] The speed limit of the autonomous vehicle is determined based on the relationship between relative speed and a preset safe speed threshold, and the relationship between relative distance and a preset safe distance threshold.

[0059] A first ST map is generated using the speed limit value, time domain, and distance domain, and a first prohibited area is marked on the first ST map;

[0060] The first ST graph is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space. The path from the starting point to the target point bypasses the first forbidden area.

[0061] The path from the starting point to the target point is smoothed to obtain a speed-planned path.

[0062] Specifically, the speed limit for the autonomous vehicle is first determined. This speed limit is then used as the slope of the straight line in the first ST graph. The first ST graph is generated using the speed limit, the time domain, and the distance domain, where S represents distance and T represents time. During the generation of the first ST graph, information obtained in the obstacle prediction phase is converted into a graphical representation. Within a specific time period Δt, if a predicted moving obstacle might occupy a distance ΔS, this portion will be marked in the first ST graph as a prohibited area.

[0063] Further, the first ST graph is discretized to obtain a mesh graph. Then, a search algorithm (such as A* algorithm or Dijkstra's algorithm) is used to find a path from the starting point to the target point. This path avoids the first forbidden region in the ST graph to prevent collisions with predicted obstacles.

[0064] Furthermore, after the ST graph sampling and search steps, the resulting path may be tortuous and discrete, which does not meet the needs of actual driving. Therefore, speed smoothing processing is required, such as using spline interpolation, to make the generated speed plan more in line with the smoothness, comfort, and handling requirements of autonomous vehicles, thus obtaining the final speed planning path.

[0065] In some embodiments, the speed limit of the autonomous vehicle is determined based on the relationship between relative speed and a preset safe speed threshold, and the relationship between relative distance and a preset safe distance threshold, including:

[0066] The relative speed is obtained by subtracting the speed of the obstacle in the detection area from the speed of the autonomous vehicle, and then comparing the relative speed with a preset safe speed threshold.

[0067] When the relative speed exceeds the safe speed threshold, execute the speed planning and control of the autonomous vehicle for the next cycle.

[0068] Otherwise, the position of the obstacle in the detection area is subtracted from the position of the autonomous vehicle to obtain the relative distance, and the relative distance is compared with the preset safe distance threshold;

[0069] When the relative distance is greater than the safe distance threshold, execute the speed planning and control of the autonomous vehicle for the next cycle; otherwise, determine the speed limit value of the autonomous vehicle.

[0070] Specifically, this application employs various methods to achieve speed planning and control. For example, a speed planning method based on time intervals or safe distances (i.e., the first speed planning method) is used. This speed planning method can be used to maintain a safe distance or safe collision time by easing braking or limiting speed when the distance between the autonomous vehicle and an obstacle is relatively short. Embodiments of this method include the calculation of forward and lateral collision times and safe distances; commonly used calculation parameters include time to collision (TTC) and following distance (THW). In practical applications, in addition to controlling the autonomous vehicle to ease braking and limit speed by calculating relative speed and relative distance, the relative angle between the obstacle and the autonomous vehicle's travel direction (i.e., the angle of the speed direction) can also be used to control the autonomous vehicle to ease braking and limit speed, thereby maintaining a safe distance or safe collision time between the autonomous vehicle and the obstacle, thus achieving more precise speed planning and control.

[0071] In one example, for instance: Figure 7 In the forward collision detection area (check_area_front) shown, assuming that obs1, obs2, obs3, and obs4 are all potentially hazardous obstacles (obs_potential_risk), then the second velocity planning method (i.e., a velocity planning method based on dynamic programming) is used to process obs1 and obs2; while the first velocity planning method (i.e., a velocity planning method based on time intervals or safe distances) is used to process obs3 and obs4. Similarly, in Figure 8 In the side collision detection area (check_area_side) shown, assuming that obs1, obs2, obs3, and obs4 are all potentially risky obstacles (obs_potential_risk), then the second velocity planning method is applied to obs4; while the first velocity planning method is applied to obs1, obs2, and obs3. Therefore, this embodiment of the application implements the logic for selecting and switching velocity planning methods based on whether there are obstacles in the detection area and the type of obstacles.

[0072] When dealing with potentially hazardous obstacles, the choice of different speed planning methods depends primarily on the relative speed and relative distance between the autonomous vehicle and the obstacle. If the relative speed exceeds a preset safe speed threshold, or the relative distance exceeds a preset safe distance threshold, the vehicle speed planning control will continue for the next cycle. Conversely, if neither of these conditions is met, this embodiment will employ gentle braking and speed limiting control to maintain a safe distance or safe collision time between the vehicle and obstacles within the detection area; that is, the first speed planning method described above (speed planning method based on time interval or safe distance) will be used for speed planning control of the autonomous vehicle.

[0073] In some embodiments, the autonomous vehicle is subjected to speed planning control using a predetermined second speed planning method, including:

[0074] By using the current state of obstacles within the detection area and the prediction model, the future state of obstacles within the detection area is predicted, and the time and distance range information of the collision between the obstacle and the autonomous vehicle's expected path are calculated based on the future state.

[0075] A second ST map is generated based on the range information of time and distance, and a second prohibited area is marked in the second ST map. The second prohibited area is used to characterize the path distance occupied by the obstacle on the expected path of the autonomous vehicle within a specific time period.

[0076] The second ST graph is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space. The path from the starting point to the target point bypasses the second forbidden region.

[0077] The path from the starting point to the target point is smoothed to obtain a speed-planned path.

[0078] Specifically, dynamic programming (DP) is an optimization algorithm used to solve optimization problems with a specific structure, typically involving the decomposition of the problem and the recursive solving of subproblems. It is applicable to many different problems, including the speed planning problem for autonomous vehicles. When using dynamic programming to plan and control the speed of an autonomous vehicle, the algorithm may include the following steps:

[0079] 1) Obstacle Prediction and Handling: This step involves environmental perception and understanding. Based on data collected by various sensors (such as radar, lidar, cameras, etc.), the future state of moving obstacles can be predicted. This is based on the obstacle's current state (such as position, velocity, direction, etc.) and a certain prediction model (such as a constant velocity model, constant acceleration model, etc.). Once the future state of the obstacle is predicted, the time and distance range at which the obstacle and the autonomous vehicle are likely to collide can be calculated.

[0080] 2) ST Map Generation: The ST map is a time-distance map, where S represents distance and T represents time. When generating the second ST map, the information obtained during the obstacle prediction phase is converted into a graphical representation. Within a specific time period Δt, if a predicted moving obstacle might occupy a distance ΔS, this portion will be marked in the second ST map as a prohibited area.

[0081] 3) ST Graph Sampling and Search: In this step, the second ST graph is first discretized to obtain a mesh graph. Then, a search algorithm (such as A* algorithm or Dijkstra's algorithm) is used to find a path from the starting point to the target point. This path avoids the second forbidden region in the second ST graph to prevent collisions with predicted obstacles.

[0082] 4) Speed ​​Smoothing: The paths obtained from the ST graph sampling and search steps may be tortuous and discrete, which does not meet the needs of actual driving. Therefore, speed smoothing is required, such as using spline interpolation, to make the generated speed plan more in line with the needs of autonomous vehicles for smoothness, comfort, and handling.

[0083] The above describes the basic process of unmanned vehicle speed planning and control based on dynamic programming algorithm in this application embodiment. The key to dynamic programming algorithm lies in effectively searching and selecting solutions to subproblems to obtain the globally optimal solution.

[0084] According to the technical solution provided in this application, by comprehensively considering the driving state and road environment information of the unmanned vehicle, the area surrounding the unmanned vehicle is divided into a forward collision detection area and a side collision detection area, forming a complete detection area. Simultaneously, based on the identification information of obstacles around the unmanned vehicle by the perception module, the obstacles around the unmanned vehicle are classified using a predetermined classification method, including categories of risk-free obstacles and risky obstacles. This classification considers not only the type of obstacle but also its cross-sectional area.

[0085] Furthermore, this application also performs speed planning control on the autonomous vehicle based on the relative speed and relative distance between the autonomous vehicle and obstacles within the detection area. When the relative speed or relative distance exceeds a preset safety threshold, the autonomous vehicle will execute the speed planning control for the next cycle as originally planned. When neither of these conditions is met, the autonomous vehicle will perform gentle braking and speed limiting control to maintain a safe distance or safe collision time between the autonomous vehicle and obstacles within the detection area.

[0086] Therefore, this application can effectively handle mixed traffic flow scenarios, especially overtaking scenarios with obstacles such as electric vehicles at close range, improving the success rate of speed planning and avoiding problems such as sudden braking and rear-end collisions that may occur with traditional ST diagram plus dynamic programming methods, thereby improving the smoothness and safety of the vehicle operation. In summary, this application achieves safe and effective speed planning and control of unmanned vehicles in complex mixed traffic environments by dynamically dividing the detection area, adjusting the obstacle type in a timely manner, and selecting an appropriate speed planning method based on the obstacle type within the detection area.

[0087] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0088] Figure 9 This is a schematic diagram of the unmanned vehicle speed planning and control device provided in an embodiment of this application. Figure 9 As shown, the unmanned vehicle speed planning and control device includes:

[0089] The segmentation module 901 is configured to collect driving status information and road environment information during the driving process of the unmanned vehicle, and to segment a detection area around the unmanned vehicle based on the driving status information and road environment information.

[0090] The classification module 902 is configured to classify the obstacles around the unmanned vehicle using a predetermined classification method based on the recognition information of the obstacles around the unmanned vehicle by the perception module, so as to determine the category of the obstacles.

[0091] The speed planning and control module 903 is configured to determine whether there are obstacles in the detection area. When there are obstacles in the detection area and the category of obstacles in the detection area is a risky obstacle category, the unmanned vehicle is speed planned and controlled using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and the obstacles in the detection area; otherwise, the unmanned vehicle is speed planned and controlled using a predetermined second speed planning method.

[0092] In some embodiments, Figure 9The segmentation module 901 divides the front collision detection area and the side collision detection area in front of and on both sides of the unmanned vehicle based on the driving status information and road environment information of the unmanned vehicle. The front collision detection area and the side collision detection area are superimposed to form a complete detection area.

[0093] In some embodiments, Figure 9 The shape and size of the front collision detection area and the side collision detection area in the partitioning module 901 change with the changes in the driving status information and road environment information of the autonomous vehicle. The default shape of the front collision detection area and the side collision detection area is a rectangle.

[0094] In some embodiments, Figure 9 The classification module 902 acquires the recognition information obtained by the perception module of the autonomous vehicle from the detection and identification of obstacles around the autonomous vehicle. The recognition information includes the obstacle type and the size of the obstacle. Based on the obstacle type and the size of the obstacle, the obstacle around the autonomous vehicle is classified. The obstacle categories include non-risk obstacle category and risk obstacle category.

[0095] In some embodiments, Figure 9 The classification module 902 determines whether an obstacle belongs to a motor vehicle or a non-motor vehicle based on the obstacle type. When the obstacle is determined to be a non-motor vehicle, or the size of the obstacle is smaller than the preset conditions, the obstacle is classified as a risky obstacle; otherwise, the obstacle is classified as a non-risky obstacle.

[0096] In some embodiments, Figure 9 The speed planning and control module 903 determines the speed limit of the unmanned vehicle based on the relationship between relative speed and a preset safe speed threshold, and the relationship between relative distance and a preset safe distance threshold. It then generates a first ST map using the speed limit, time domain, and distance domain, marking a first prohibited area on the first ST map. The first ST map is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space, bypassing the first prohibited area. Finally, the path from the starting point to the target point is speed-smoothed to obtain a speed-smoothed planned path.

[0097] In some embodiments, Figure 9The speed planning and control module 903 subtracts the speed of the obstacle in the detection area from the speed of the unmanned vehicle to obtain the relative speed, and compares the relative speed with a preset safe speed threshold. If the relative speed is greater than the safe speed threshold, the unmanned vehicle speed planning and control for the next cycle is executed; otherwise, the position of the obstacle in the detection area is subtracted from the position of the unmanned vehicle to obtain the relative distance, and the relative distance is compared with a preset safe distance threshold. If the relative distance is greater than the safe distance threshold, the unmanned vehicle speed planning and control for the next cycle is executed; otherwise, the speed limit value of the unmanned vehicle is determined.

[0098] In some embodiments, Figure 9 The speed planning and control module 903 uses the current state and prediction model of obstacles within the detection area to predict the future state of obstacles within the detection area. Based on the future state, it calculates the time and distance range information of the collision between the obstacle and the expected path of the autonomous vehicle. A second ST map is generated based on the time and distance range information, and a second prohibited area is marked on the second ST map. The second prohibited area represents the path distance occupied by the obstacle on the expected path of the autonomous vehicle within a specific time period. The second ST map is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space. The path from the starting point to the target point bypasses the second prohibited area. The path from the starting point to the target point is then speed-smoothed to obtain the speed-smoothed planned path.

[0099] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0100] Figure 10 This is a schematic diagram of the structure of the electronic device 10 provided in an embodiment of this application. Figure 10 As shown, the electronic device 10 of this embodiment includes: a processor 1001, a memory 1002, and a computer program 1003 stored in the memory 1002 and executable on the processor 1001. When the processor 1001 executes the computer program 1003, it implements the steps in the various method embodiments described above. Alternatively, when the processor 1001 executes the computer program 1003, it implements the functions of each module / unit in the various device embodiments described above.

[0101] For example, computer program 1003 may be divided into one or more modules / units, which are stored in memory 1002 and executed by processor 1001 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 1003 in electronic device 10.

[0102] Electronic device 10 may be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 10 may include, but is not limited to, a processor 1001 and a memory 1002. Those skilled in the art will understand that... Figure 10 This is merely an example of electronic device 10 and does not constitute a limitation on electronic device 10. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device may also include input / output devices, network access devices, buses, etc.

[0103] The processor 1001 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0104] The memory 1002 can be an internal storage unit of the electronic device 10, such as a hard disk or RAM of the electronic device 10. The memory 1002 can also be an external storage device of the electronic device 10, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, FlashCard, etc., equipped on the electronic device 10. Furthermore, the memory 1002 can include both internal and external storage units of the electronic device 10. The memory 1002 is used to store computer programs and other programs and data required by the electronic device. The memory 1002 can also be used to temporarily store data that has been output or will be output.

[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0106] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0107] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments claimed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0108] In the embodiments provided in this application, it should be understood that the disclosed apparatus / computer devices and methods can be implemented in other ways. For example, the apparatus / computer device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0109] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0110] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0111] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added to or subtracted according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0112] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A speed planning and control method for an unmanned vehicle, characterized in that, include: The driving status information and road environment information of the unmanned vehicle are collected during the driving process. Based on the driving status information and road environment information, a detection area is delineated around the unmanned vehicle. Based on the identification information of obstacles around the unmanned vehicle by the perception module, the obstacles around the unmanned vehicle are classified using a predetermined classification method to determine the category of the obstacles; Determine whether there are obstacles in the detection area. If there are obstacles in the detection area and the category of obstacles in the detection area is a risky obstacle category, perform speed planning control on the unmanned vehicle using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and the obstacles in the detection area; otherwise, perform speed planning control on the unmanned vehicle using a predetermined second speed planning method. The step of performing speed planning control on the unmanned vehicle using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and obstacles within the detection area includes: Based on the relationship between the relative speed and a preset safe speed threshold, and the relationship between the relative distance and a preset safe distance threshold, the speed limit of the unmanned vehicle is determined; a first ST map is generated using the speed limit, the time domain, and the distance domain, and a first prohibited area is marked in the first ST map; the first ST map is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space, the path from the starting point to the target point bypassing the first prohibited area; the path from the starting point to the target point is subjected to speed smoothing processing to obtain the speed-smoothed speed planning path.

2. The method according to claim 1, characterized in that, The step of defining a detection zone around the unmanned vehicle based on the driving status information and road environment information includes: Based on the driving status information and road environment information of the unmanned vehicle, a frontal collision detection area and a side collision detection area are respectively defined in front of and on both sides of the unmanned vehicle. The frontal collision detection area and the side collision detection area are superimposed to form a complete detection area.

3. The method according to claim 2, characterized in that, The method further includes: The shape and size of the forward collision detection area and the side collision detection area change with the changes in the driving status information and the road environment information of the autonomous vehicle. The default shape of the forward collision detection area and the side collision detection area is a rectangle.

4. The method according to claim 1, characterized in that, The step of classifying the obstacles around the autonomous vehicle using a predetermined classification method based on the obstacle recognition information obtained by the perception module includes: The identification information obtained by the perception module of the unmanned vehicle through the detection and identification of obstacles around the unmanned vehicle includes the type and size of the obstacle. The obstacles around the autonomous vehicle are classified according to their type and size, and the categories of obstacles include risk-free obstacles and risky obstacles.

5. The method according to claim 4, characterized in that, The classification of obstacles around the unmanned vehicle based on the type and size of the obstacles includes: Based on the type of obstacle, determine whether the obstacle belongs to a motor vehicle or a non-motor vehicle. If the obstacle is determined to be a non-motor vehicle, or if the size of the obstacle is smaller than a preset condition, the obstacle is classified as a risky obstacle; otherwise, the obstacle is classified as a non-risky obstacle.

6. The method according to claim 1, characterized in that, The step of determining the speed limit of the autonomous vehicle based on the relationship between the relative speed and a preset safe speed threshold, and the relationship between the relative distance and a preset safe distance threshold, includes: The relative speed is obtained by subtracting the speed of the obstacle in the detection area from the speed of the unmanned vehicle, and then comparing the relative speed with a preset safe speed threshold. When the relative speed is greater than the safe speed threshold, the autonomous vehicle speed planning and control for the next cycle is executed; Otherwise, the relative distance is obtained by subtracting the position of the obstacle in the detection area from the position of the unmanned vehicle, and the relative distance is compared with a preset safe distance threshold. When the relative distance is greater than the safe distance threshold, the speed planning and control of the autonomous vehicle in the next cycle is executed; otherwise, the speed limit value of the autonomous vehicle is determined.

7. The method according to claim 1, characterized in that, The speed planning and control of the unmanned vehicle using a predetermined second speed planning method includes: Using the current state of obstacles within the detection area and a prediction model, the future state of obstacles within the detection area is predicted, and based on the future state, the time and distance range information of the collision between the obstacle and the expected path of the unmanned vehicle are calculated. A second ST map is generated based on the time and distance range information, and a second prohibited area is marked in the second ST map. The second prohibited area is used to characterize the path distance occupied by the obstacle on the expected path of the unmanned vehicle within a specific time period. The second ST graph is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space, wherein the path from the starting point to the target point bypasses the second forbidden region. The path from the starting point to the target point is subjected to speed smoothing processing to obtain the speed-smoothed planned path.

8. A speed planning and control device for an unmanned vehicle, characterized in that, include: The segmentation module is configured to collect driving status information and road environment information during the autonomous vehicle's operation, and to segment a detection area around the autonomous vehicle based on the driving status information and road environment information. The classification module is configured to classify the obstacles around the unmanned vehicle using a predetermined classification method based on the identification information of the obstacles around the unmanned vehicle by the perception module, so as to determine the category of the obstacles; The speed planning and control module is configured to determine whether there are obstacles in the detection area. When there are obstacles in the detection area and the category of obstacles in the detection area is a risky obstacle category, the unmanned vehicle is speed planned and controlled using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and the obstacles in the detection area; otherwise, the unmanned vehicle is speed planned and controlled using a predetermined second speed planning method. The step of performing speed planning control on the unmanned vehicle using a predetermined first speed planning method based on the relative speed and relative distance between the unmanned vehicle and obstacles within the detection area includes: Based on the relationship between the relative speed and a preset safe speed threshold, and the relationship between the relative distance and a preset safe distance threshold, the speed limit of the unmanned vehicle is determined; a first ST map is generated using the speed limit, the time domain, and the distance domain, and a first prohibited area is marked in the first ST map; the first ST map is discretized, and for each target point, a search algorithm is used to search for a path from the starting point to the target point in the ST search space, the path from the starting point to the target point bypassing the first prohibited area; the path from the starting point to the target point is subjected to speed smoothing processing to obtain the speed-smoothed speed planning path.

9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images