Automatic driving path planning method and readable storage medium thereof

Abstract

The invention belongs to the technical field of automatic driving, and particularly relates to an automatic driving path planning method, which comprises the steps of S1, environment information acquisition and preprocessing: acquiring environment information around a vehicle by using various sensors, and performing fusion processing on the acquired multi-source sensor information, S2, map construction and updating, and S3, automatic driving path planning. The method comprises the following steps: S1, pre-processing environment information of a vehicle, constructing a local map according to the pre-processed environment information, and updating the map in real time according to received traffic information, S3, searching and optimizing a path, searching an initial path from a current position of the vehicle to a target position based on the constructed and updated map, and optimizing the searched initial path; the accuracy and integrity of the environment information are improved, so that the path can be planned more accurately, the flexibility and adaptability of path planning are improved, the operation cost of the vehicle is reduced, the driving efficiency is improved, and the safety and the driving continuity of the automatic driving vehicle are guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous driving, and in particular relates to an autonomous driving path planning method and a readable storage medium thereof. Background Art

[0002] In autonomous driving, the accuracy of the map will directly affect the decision-making ability of the autonomous driving system, and then directly affect the control of the smart car, thus affecting the safety and comfort of smart driving. It can be seen that for the autonomous driving system, clear map data can clearly locate the position of the vehicle, and then better plan a reasonable and smooth driving trajectory, which plays a key role in the safe and stable control of the smart vehicle. Therefore, providing high-precision map data is the key to the autonomous driving system, which can effectively improve the driving safety of autonomous driving vehicles and enhance the overall perception and decision-making capabilities of the autonomous driving system;

[0003] Existing autonomous driving path planning solutions have some limitations, such as insufficient adaptability to complex terrain and special road conditions, the difficulty in ensuring the efficiency and real-time performance of path planning when faced with large amounts of traffic data, and the lack of flexibility in the optimization objectives of path planning under different driving scenarios.

[0004] To this end, an autonomous driving path planning method and a readable storage medium thereof are designed to solve the above problems. Summary of the invention

[0005] To solve the problems raised in the above background technology, the present invention provides an automatic driving path planning method and a readable storage medium thereof, which can effectively solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: an automatic driving path planning method, the automatic driving path planning method comprising:

[0007] S1: Environmental information collection and preprocessing, using multiple sensors to collect environmental information around the vehicle, and fusing the collected multi-source sensor information;

[0008] S2: Map construction and update: build a local map based on the preprocessed environmental information and update the map in real time based on the received traffic information;

[0009] S3: Path search and optimization: Based on the constructed and updated map, the initial path from the vehicle's current location to the target location is searched, and the searched initial path is optimized;

[0010] S4: Real-time dynamic route adjustment, including dynamic monitoring of the local environment, route reassessment and adjustment, and interactive adjustment of traffic information.

[0011] As an automatic driving path planning method of the present invention, in the environmental information collection and preprocessing steps, the multiple sensors include lidar, camera and millimeter wave radar, and the Kalman filtering algorithm is used for data fusion.

[0012] As an automatic driving path planning method of the present invention, in the map construction and update step, the local map is constructed in the form of a grid map or a topological map.

[0013] As an automatic driving path planning method of the present invention, in the path search and optimization step, the A* algorithm or the Dijkstra algorithm is used to perform an initial path search, and a cost function is used to optimize the searched initial path, and the cost function comprehensively considers the path length, driving time and energy consumption.

[0014] As an automatic driving path planning method of the present invention, in the real-time dynamic path adjustment step, the local environment is dynamically monitored by laser radar, millimeter wave radar and camera to monitor environmental changes in real time, the path re-evaluation and adjustment uses ant colony algorithm to find alternative paths and evaluate them according to the planning target priority, and the traffic information interactive adjustment receives information and adjusts the path planning through vehicle-to-vehicle and vehicle-to-road communication technology.

[0015] The technical solution also provides a readable storage medium, which includes a program code for implementing an autonomous driving path planning method and pre-collected map data, and adopts a hierarchical storage structure and an index mechanism to improve access efficiency.

[0016] As a readable storage medium of the present invention, the storage medium is a hard disk or a flash memory.

[0017] Compared with the prior art, the invention has the following beneficial effects: the invention has a scientific and reasonable structure and is safe and convenient to use:

[0018] 1. The automatic driving path planning method of the present invention collects environmental information through multi-sensor fusion, thereby improving the accuracy and completeness of environmental information, thereby being able to plan the path more accurately.

[0019] 2. The real-time construction and update mechanism of the map enables route planning to adapt to changes in the traffic environment, such as road construction, traffic sign changes, etc., improving the flexibility and adaptability of route planning.

[0020] 3. The path search and optimization process comprehensively considers multiple factors, such as path length, driving time and energy consumption, to obtain a better path, reduce vehicle operating costs and improve driving efficiency.

[0021] 4. The real-time dynamic path adjustment mechanism can respond to various emergencies and changes in traffic information in a timely manner to ensure the safety and driving continuity of autonomous driving vehicles.

[0022] 5. The readable storage medium adopts a reasonable storage structure and index mechanism, which improves the access efficiency of program code and map data, and helps to improve the real-time performance of the entire autonomous driving path planning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0024] Figure 1 Principle block diagram of the present invention; DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Example: Figure 1 As shown, the present invention provides a technical solution, an automatic driving path planning method, and the automatic driving path planning method includes:

[0027] S1: Environmental information collection and preprocessing, using multiple sensors to collect environmental information around the vehicle, and fusing the collected multi-source sensor information;

[0028] S2: Map construction and update: build a local map based on the preprocessed environmental information and update the map in real time based on the received traffic information;

[0029] S3: Path search and optimization: Based on the constructed and updated map, the initial path from the vehicle's current location to the target location is searched, and the searched initial path is optimized;

[0030] S4: Real-time dynamic route adjustment, including dynamic monitoring of the local environment, route reassessment and adjustment, and interactive adjustment of traffic information.

[0031] In an optional embodiment, in the environmental information collection and preprocessing step, multiple sensors include lidar, camera and millimeter wave radar, and a Kalman filter algorithm is used for data fusion.

[0032] In this implementation scheme:

[0033] 1. Sensor installation and configuration: Install sensors such as LiDAR, cameras, and millimeter-wave radar on autonomous vehicles. The LiDAR is installed on the top of the vehicle to ensure that its scanning range can cover most of the front, sides, and rear of the vehicle. Cameras are installed on the front windshield, rearview mirror, and other locations of the vehicle to obtain visual images from different perspectives. Millimeter-wave radars are installed on the front and rear bumpers of the vehicle, focusing on monitoring dynamic objects around the vehicle, and configuring parameters according to the characteristics of the sensor. For example, set the scanning frequency, resolution, and other parameters of the LiDAR, the focal length, resolution, frame rate, and other parameters of the camera, and the detection range, frequency, and other parameters of the millimeter-wave radar to ensure that high-quality environmental information is collected.

[0034] 2. Data fusion is implemented by running the data fusion program in the vehicle's computing unit (such as an onboard computer). First, the hardware clock synchronization circuit is used to synchronize the data from different sensors. Then, the point cloud data collected by the lidar, the image data collected by the camera, and the distance and speed data collected by the millimeter-wave radar are input into the Kalman filter algorithm module. The Kalman filter algorithm performs weighted fusion on the data based on the uncertainty of each sensor data (such as the decrease in the accuracy of the lidar data at long distances, the influence of light on the camera image, etc.), and outputs the fused environmental information, including the precise position, shape, speed and other information of objects around the vehicle.

[0035] In an optional embodiment, in the map construction and update step, the local map is constructed in the form of a grid map or a topological map.

[0036] In this implementation scheme:

[0037] 1. Local map construction

[0038] When the vehicle starts, it starts to build a local map based on the fused environmental information. If the grid map construction method is used, the area around the vehicle is divided into grids of equal size (for example, a square grid with a side length of 0.5 meters). According to the object information detected by the sensor, the grid is marked as different states, such as road grid, obstacle grid, unknown grid, etc.

[0039] If a topological map is constructed, key nodes in the environment (such as road intersections, lane merging points, etc.) and edges connecting these nodes (such as road sections) are identified, and attributes (such as length, speed limit, etc.) are assigned to the edges to construct a topological map around the vehicle.

[0040] 2. Map update operation

[0041] Vehicles receive real-time traffic information through on-board communication devices (such as vehicle networking modules). When road construction information is received, the grid at the corresponding position in the grid map is marked as a construction area (for example, marked with a special color) through a map update program, and the corresponding road section is marked as impassable in the topological map.

[0042] When new traffic sign information is received, the traffic sign identification is added or updated at the corresponding position in the map. For example, if a change in the speed limit sign ahead is received, the speed limit value is updated at the corresponding road section in the map.

[0043] In an optional embodiment, in the path search and optimization step, the A* algorithm or the Dijkstra algorithm is used to perform an initial path search, and a cost function is used to optimize the searched initial path, wherein the cost function comprehensively considers the path length, driving time, and energy consumption.

[0044] In this implementation scheme:

[0045] 1. Initial path search

[0046] Based on the constructed map, the A* algorithm is used to perform the initial path search, with the vehicle's current position as the starting node and the target position (such as the coordinates of the destination or a predefined parking point) as the target node.

[0047] During the search process, the estimated cost from each node to the target node (such as using the Euclidean distance as a heuristic function) and the actual cost from the starting node to the node (taking into account factors such as road conditions and traffic rules) are calculated. The next expansion node is selected based on the total cost until the target node is found and the initial path is obtained.

[0048] 2. Path Optimization Process

[0049] For the initial path searched, construct a cost function. Let the weight of the path length be, the weight of the travel time be, and the weight of the energy consumption be, and the cost function is, where is the path length, is the travel time, and is the energy consumption.

[0050] The cost is optimized by adjusting the nodes in the path. For example, if a certain section of road is found to be congested, the driving time can be reduced by changing lanes or choosing other parallel roads, thereby reducing the cost. The optimization is iterated continuously until the cost no longer decreases and the optimal path is obtained.

[0051] It should be noted that A* is an effective algorithm for finding the shortest path on a graph plane. The A* algorithm combines the advantages of the greedy algorithm and the Dijkstra algorithm. It uses a heuristic function to evaluate the distance from the current node to the target node, thereby guiding the search direction and finding the optimal path faster.

[0052] Its principle is 1. Maintain two lists: open list and closed list. The open list stores the nodes to be examined, and the closed list stores the nodes that have been examined. 2. Start from the starting node and put it into the open list. Calculate the estimated distance from the starting node to the target node (heuristic function value) and the actual distance from the starting node to itself (usually 0) to get the total cost of the starting node. 3. Select the node with the smallest total cost from the open list for expansion. Remove the node from the open list and put it into the closed list. 4. Process the adjacent nodes of the expanded node: if the adjacent node is in the closed list, ignore it; if the adjacent node is not in the open list, calculate its total cost and add it to the open list; if the adjacent node is already in the open list, compare the new total cost with the old total cost. If the new total cost is smaller, update the total cost and parent node information of the node. 5. Repeat steps 3 and 4 until the target node is added to the open list or the open list is empty. If the target node is added to the open list, the shortest path from the starting node to the target node can be found by backtracking the parent node.

[0053] The A* algorithm is widely used in many fields, such as character pathfinding in game development, robot path planning, map navigation, etc.

[0054] In an optional embodiment, in the real-time dynamic path adjustment step, the local environment is dynamically monitored by lidar, millimeter wave radar and camera to monitor environmental changes in real time, the path re-evaluation and adjustment uses ant colony algorithm to find alternative paths and evaluate them according to the planning target priority, and the traffic information interaction adjustment receives information and adjusts the path planning through vehicle-to-vehicle and vehicle-to-road communication technology.

[0055] In this implementation scheme:

[0056] 1. Implementation of dynamic monitoring of local environment

[0057] During the driving process of the vehicle, LiDAR, millimeter-wave radar and cameras continuously collect data. LiDAR data is used to update the micro-environment information around the vehicle in real time, especially the location and movement speed of obstacles. Millimeter-wave radar data is used to monitor the speed and distance changes of moving targets around the vehicle, and camera data is used to identify changes in traffic signs (such as temporary speed limit signs) and road construction signs.

[0058] These sensor data are fused and processed in real time, and once new situations are detected (such as new obstacles, changes in traffic signs or road construction), the path re-evaluation mechanism is immediately triggered.

[0059] 2. Path re-evaluation and adjustment implementation

[0060] When the path re-evaluation mechanism is triggered, the ant colony algorithm is used to find an alternative path nearby based on the current meso-environmental model and micro-environmental information, with the current vehicle location as the starting point and the destination as the end point. The pheromone update in the ant colony algorithm is dynamically adjusted according to factors such as the traffic conditions of the road section and the road type.

[0061] After finding an alternative path, the alternative path is evaluated according to the planning target priority in the current driving scenario. If the alternative path meets the priority requirements, a smooth transition plan is calculated to switch from the current driving path to the alternative path. Through the coordinated control of the vehicle's power system, steering system, etc., a gradual change in speed and direction is achieved to ensure smooth path switching.

[0062] 3. Implementation of traffic information interaction adjustment

[0063] The vehicle exchanges information with surrounding vehicles through vehicle-to-vehicle communication technology. For example, it broadcasts its position, speed, driving direction and other information to surrounding vehicles at regular intervals (such as 1 second), and receives similar information sent by surrounding vehicles. If the driving intention of the vehicle in front (such as lane change intention) is received, it will make adjustments based on the current path planning situation. If the lane change of the vehicle in front may affect the safety or efficiency of the current path, the path will be re-evaluated and adjusted.

[0064] Vehicles exchange information with roadside infrastructure through vehicle-road communication technology. When approaching roadside infrastructure (such as traffic lights and variable message signs), they receive traffic control information (such as temporary speed limits and lane change instructions) sent by the infrastructure. Based on this information, the route planning is adjusted in a timely manner to ensure that the vehicle complies with traffic regulations and drives efficiently.

[0065] The technical solution also provides a readable storage medium, including a program code for implementing an autonomous driving path planning method, and pre-collected map data, and adopts a hierarchical storage structure and index mechanism to improve access efficiency. The storage medium is a hard disk or a flash memory.

[0066] In this implementation scheme:

[0067] 1. Storage content

[0068] The storage medium (such as a hard disk, flash memory, etc.) is used to store the program code for implementing the above-mentioned automatic driving path planning method. The program code includes module codes for sensor data acquisition and fusion, module codes for map construction and updating, module codes for path search and optimization, and module codes for dynamic path adjustment.

[0069] At the same time, the storage medium also stores pre-collected map data (such as high-precision urban road maps, highway maps, etc.), which can be used as basic data for path planning to reduce the amount of calculation for real-time map construction.

[0070] 2. Storage structure and access efficiency

[0071] A hierarchical storage structure is used to store program code and map data. For example, frequently accessed map data and program modules are stored in the cache layer to increase access speed. For program code, it is classified and stored according to functional modules to facilitate quick positioning and loading of required modules.

[0072] An indexing mechanism is established to improve the efficiency of access to stored content. For example, indexes are established for different areas in the map data (such as maps of different cities). When a vehicle enters a specific area, the map data and related program code of that area can be quickly located and loaded.

[0073] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An automatic driving path planning method, characterized in that: The automatic driving path planning method comprises: S1: Environmental information collection and preprocessing, using multiple sensors to collect environmental information around the vehicle, and fusing the collected multi-source sensor information; S2: Map construction and update: build a local map based on the preprocessed environmental information and update the map in real time based on the received traffic information; S3: Path search and optimization: Based on the constructed and updated map, the initial path from the vehicle's current location to the target location is searched, and the searched initial path is optimized; S4: Real-time dynamic route adjustment, including dynamic monitoring of the local environment, route reassessment and adjustment, and interactive adjustment of traffic information.

2. The method for autonomous driving path planning according to claim 1, characterized in that: In the environmental information collection and preprocessing step, the multiple sensors include laser radar, camera and millimeter wave radar, and the Kalman filter algorithm is used for data fusion.

3. The method for autonomous driving path planning according to claim 1, characterized in that: In the map construction and updating step, the local map is constructed in the form of a grid map or a topological map.

4. The method for autonomous driving path planning according to claim 1, characterized in that: In the path search and optimization step, the A* algorithm or the Dijkstra algorithm is used to perform an initial path search, and a cost function is used to optimize the searched initial path, wherein the cost function comprehensively considers the path length, travel time and energy consumption.

5. The method for autonomous driving path planning according to claim 1, wherein: In the real-time dynamic path adjustment step, the local environment is dynamically monitored through lidar, millimeter wave radar and camera to monitor environmental changes in real time, the path re-evaluation and adjustment uses ant colony algorithm to find alternative paths and evaluate them according to the planning target priority, and the traffic information interactive adjustment receives information and adjusts the path planning through vehicle-to-vehicle and vehicle-to-road communication technology.

6. A readable storage medium, characterized in that: The system stores program code for implementing the autonomous driving path planning method described in any one of claims 1 to 5, as well as pre-collected map data, and adopts a hierarchical storage structure and an index mechanism to improve access efficiency.

7. A readable storage medium according to claim 6, characterized in that: The storage medium is a hard disk or a flash memory.

Images