Vehicle driving path planning method

Abstract

The invention discloses a vehicle driving path planning method, and relates to the technical field of data fusion and path planning. Comprising the following steps: step 1: carrying out data acquisition and preprocessing, step 2: carrying out environment modeling according to preprocessed road network information: constructing a dynamic road network model G = (V, E), V representing nodes in a road network, E representing edges between the nodes, and a weight omega (e) of the dynamic road network model representing time cost of passing through a road section, step 3: carrying out path search, and step 4: carrying out path search. 4, performing dynamic obstacle avoidance: detecting dynamic obstacles in the surrounding environment according to sensor data, if the obstacles are detected, performing obstacle avoidance by adjusting a local path, and if the obstacles cannot be avoided by adjusting the local path, triggering global re-planning and performing path search again; and generating a vehicle control instruction according to the final path to control vehicle driving.

Description

Technical Field

[0001] The present invention discloses a vehicle driving path planning method, which relates to the technical fields of data fusion and path planning. Background Art

[0002] With the development of intelligent transportation systems, vehicle path planning technology has become a core component of autonomous driving and intelligent navigation systems. Existing path planning methods mainly rely on static map data and fixed path selection algorithms, and are unable to effectively cope with real-time traffic changes, unexpected road conditions, and user personalized needs. Existing path planning systems usually rely on shortest path algorithms, such as Dijkstra algorithm or A*

[0003] algorithm, but these methods often perform poorly in complex traffic environments, especially in traffic jams, accident-prone areas, or bad weather conditions, unable to handle complex traffic environments, and unable to meet users' needs for efficient, safe, and personalized paths. Summary of the Invention

[0004] In view of the problems of the existing technology, the present invention provides a vehicle driving path planning method, which solves the deficiencies of traditional path planning in terms of real-time performance, safety, and multi-objective optimization through dynamic data fusion, intelligent algorithm optimization, and scenario adaptability design, thereby improving the driving efficiency and safety of vehicles in different environments.

[0005] The specific solution proposed by the present invention is as follows:

[0006] The present invention provides a vehicle driving path planning method, including:

[0007] Step 1: Perform data collection and preprocessing:

[0008] Collect multi-source data, where the multi-source data includes GPS data, sensor data, and road network information,

[0009] Perform preprocessing on the collected data,

[0010] Step 2: Perform environmental modeling based on the preprocessed road network information:

[0011] Construct a dynamic road network model G=(V, E), where V represents the nodes in the road network, E represents the edges between the nodes, and the weight ω(e) of the dynamic road network model represents the time cost of passing through the road section,

[0012] Step 3: Perform path search:

[0013] Taking the shortest time as the optimization goal, use an improved A* algorithm to perform path search, and make the formula of the heuristic function h(n) as:

[0014]

[0015] where S max is the maximum allowable driving speed in the road network, (x goal - y goal ) is the target location, (x n - y n ) is the current location.

[0016] Then the cost function f(n) = g(n) + h(n), where g(n) is the actual time cost from the starting point to node n.

[0017] During the search process, update the weight ω(e) of the dynamic road network model.

[0018] Step 4: Perform dynamic obstacle avoidance:

[0019] Detect dynamic obstacles in the surrounding environment according to the sensor data. If an obstacle is detected, adjust the local path for obstacle avoidance. If the local path adjustment cannot avoid the obstacle, trigger global replanning and re - perform path search.

[0020] Step 5: Output the final path and generate vehicle control instructions according to the final path to control the vehicle to drive.

[0021] Furthermore, the GPS data collected in step 1 of the vehicle driving path planning method described above includes the vehicle's current location, speed, and direction.

[0022] The sensor data collected includes the surrounding environment information collected by lidar and cameras.

[0023] The road network information collected includes road topology, lane information, traffic rules, and real - time traffic flow data.

[0024] Furthermore, in step 1 of the vehicle driving path planning method described above, pre - process the collected data, including: data cleaning, aligning the data coordinates, and synchronizing the data time.

[0025] Furthermore, in step 2 of the vehicle driving path planning method described above, use the formula to calculate ω(e), where L(e) is the road segment length, S(e) is the average driving speed of the road segment, and T delay (e) is the road segment delay time.

[0026] The present invention also provides a vehicle driving path planning device, including a data collection and pre - processing module, a modeling module, a search module, an obstacle avoidance module, and an output module.

[0027] The data collection and pre - processing module performs data collection and pre - processing:

[0028] Collect multi-source data, where the multi-source data includes GPS data, sensor data, and road network information.

[0029] Preprocess the collected data.

[0030] The modeling module performs environmental modeling based on the preprocessed road network information:

[0031] Construct a dynamic road network model G=(V, E), where V represents the nodes in the road network, E represents the edges between the nodes, and the weight ω(e) of the dynamic road network model represents the time cost of passing through the road section.

[0032] The search module conducts path search:

[0033] With the shortest time as the optimization goal, use an improved A* algorithm for path search, and the formula for the heuristic function h(n) is:

[0034]

[0035] Where S max is the maximum allowable driving speed in the road network, (x goal -y goal ) is the target location, and (x n -y n ) is the current location.

[0036] Then the cost function f(n)=g(n)+h(n), where g(n) is the actual time cost from the starting point to node n.

[0037] During the search process, update the weight ω(e) of the dynamic road network model.

[0038] The obstacle avoidance module conducts dynamic obstacle avoidance:

[0039] Detect dynamic obstacles in the surrounding environment according to the sensor data. If an obstacle is detected, adjust the local path for obstacle avoidance. If the local path adjustment cannot avoid the obstacle, trigger global replanning and re-conduct path search.

[0040] The output module outputs the final path and generates vehicle control instructions according to the final path to control the vehicle to drive.

[0041] Furthermore, the GPS data collected by the acquisition and preprocessing module of the vehicle driving path planning device includes the vehicle's current location, speed, and direction.

[0042] The collected sensor data includes the surrounding environment information collected by lidar and cameras.

[0043] The collected road network information includes road topology, lane information, traffic rules, and real-time traffic flow data.

[0044] Furthermore, the acquisition and preprocessing module of the vehicle driving path planning device preprocesses the acquired data, including: data cleaning, aligning the data coordinates, and synchronizing the data time.

[0045] Furthermore, the modeling module of the vehicle driving path planning device uses the formula to calculate ω(e), where L(e) is the road segment length, S(e) is the average driving speed of the road segment, and T delay (e) is the road segment delay time.

[0046] The advantages of the present invention are as follows:

[0047] Through multi-source data fusion, dynamic environment modeling, improved path search algorithm and real-time obstacle avoidance strategy, the present invention provides an efficient, accurate and reliable vehicle driving path planning method, which has broad application prospects and can significantly promote the development of intelligent transportation and autonomous driving technologies. Compared with the common vehicle path planning methods at present, it has the following advantages:

[0048] Multi-source data fusion: By fusing GPS data, sensor data and road network information, the accuracy and reliability of path planning are improved.

[0049] Real-time dynamic planning: Obstacles (such as pedestrians, other vehicles) are detected in real time through sensor data, and local obstacle avoidance strategies or global replanning are adopted to ensure the safe driving of the vehicle. It can handle complex road network structures (such as multi-lanes, intersections, roundabouts, etc.) and adapt to complex scenarios.

[0050] Shortest time optimization: Aiming at the shortest time, it meets the user's demand for efficient travel. Combining real-time traffic flow data, the path planning is dynamically adjusted to avoid congested road segments and further shorten the driving time.

[0051] Strong robustness: Supports dynamic obstacle avoidance and replanning to ensure the safe driving of the vehicle. Description of the Drawings

[0052] Figure 1 is a schematic flow diagram of the method of the present invention. Detailed Embodiments

[0053] The present invention will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it, but the embodiments given are not intended to limit the present invention.

[0054] Embodiment 1

[0055] The present invention provides a vehicle driving path planning method, including:

[0056] Step 1: Conduct data collection and preprocessing:

[0057] Collect multi-source data, including GPS data, sensor data, and road network information. The collected GPS data includes the vehicle's current location, speed, and direction. The collected sensor data includes the surrounding environment information collected by lidar and cameras. The collected road network information includes road topology, lane information, traffic rules, and real-time traffic flow data.

[0058] Preprocess the collected data, including: data cleaning, aligning the data coordinates, and synchronizing the data time.

[0059] Step 2: Perform environmental modeling based on the preprocessed road network information:

[0060] Construct a dynamic road network model G=(V, E), where V represents the nodes in the road network, such as intersections, key sections, etc., and E represents the edges between nodes. The weight ω(e) of the dynamic road network model represents the time cost of passing through the section. Among them, use the formula to calculate ω(e), L(e) is the length of the section, S(e) is the average driving speed of the section, which can be dynamically adjusted according to the real-time traffic flow data, and T delay (e) is the section delay time.

[0061] Step 3: Conduct path search:

[0062] Taking the shortest time as the optimization goal, use the improved A* algorithm to conduct path search, and make the formula of the heuristic function h(n) as:

[0063]

[0064] where S max is the maximum allowable driving speed in the road network, (x goal -y goal ) is the target location, and (x n -y n ) is the current location.

[0065] Then the cost function f(n)=g(n)+h(n), where g(n) is the actual time cost from the starting point to node n.

[0066] During the search process, update the weight ω(e) of the dynamic road network model.

[0067] Step 4: Conduct dynamic obstacle avoidance:

[0068] Detect dynamic obstacles in the surrounding environment according to the sensor data. If an obstacle is detected, the dynamic window method can be used to adjust the local path for obstacle avoidance. If the local path cannot be adjusted to avoid the obstacle, then trigger global replanning and conduct path search again.

[0069] Step 5: Output the final path, and generate vehicle control instructions according to the final path to control the vehicle to drive, such as steering, accelerating, braking, etc. During the vehicle driving process, continuously monitor the environmental changes and update the path in real time.

[0070] The method of the present invention is applicable to autonomous driving vehicles, logistics distribution vehicles and shared mobility vehicles. It can support multi-vehicle collaborative path planning, and avoid conflicts and congestion by coordinating the paths of each vehicle. It can be implemented in an embedded system with limited computing resources, and reduce the computational complexity through an optimized algorithm.

[0071] Embodiment 2

[0072] The present invention also provides a vehicle driving path planning device, including a collection and preprocessing module, a modeling module, a search module, an obstacle avoidance module and an output module.

[0073] The collection and preprocessing module performs data collection and preprocessing:

[0074] Collect multi-source data, and the multi-source data includes GPS data, sensor data and road network information.

[0075] Preprocess the collected data.

[0076] The modeling module performs environmental modeling according to the preprocessed road network information:

[0077] Construct a dynamic road network model G=(V, E), where V represents the nodes in the road network, E represents the edges between the nodes, and the weight ω(e) of the dynamic road network model represents the time cost of passing through the road section.

[0078] The search module performs path search:

[0079] Taking the shortest time as the optimization goal, use an improved A* algorithm to perform path search, and make the formula of the heuristic function h(n) be:

[0080]

[0081] Where S max is the maximum driving speed allowed in the road network, (x goal -y goal ) is the target position, (x n -y n ) is the current position.

[0082] Then the cost function f(n)=g(n)+h(n), where g(n) is the actual time cost from the starting point to node n.

[0083] During the search process, update the weight ω(e) of the dynamic road network model.

[0084] The obstacle avoidance module performs dynamic obstacle avoidance:

[0085] Detect dynamic obstacles in the surrounding environment based on sensor data. If an obstacle is detected, adjust the local path to avoid it. If the local path adjustment cannot avoid the obstacle, trigger global replanning and re - conduct path search.

[0086] The output module outputs the final path and generates vehicle control instructions according to the final path to control the vehicle to drive.

[0087] Regarding the information interaction, execution process, etc. among the modules in the above - mentioned device, since they are based on the same concept as the method embodiment of the present invention, the specific content can be referred to the description in the method embodiment of the present invention, and will not be elaborated here.

[0088] Similarly, the device of the present invention provides an efficient, accurate, and reliable vehicle driving path planning method through multi - source data fusion, dynamic environment modeling, improved path search algorithms, and real - time obstacle avoidance strategies, and has broad application prospects, which can significantly promote the development of intelligent transportation and autonomous driving technologies. Compared with the common vehicle path planning methods currently, it has the following advantages:

[0089] Multi - source data fusion: By fusing GPS data, sensor data, and road network information, the accuracy and reliability of path planning are improved.

[0090] Real - time dynamic planning: Detect obstacles (such as pedestrians, other vehicles) in real time through sensor data, and adopt local obstacle avoidance strategies or global replanning to ensure the safe driving of the vehicle. It can handle complex road network structures (such as multi - lanes, intersections, roundabouts, etc.) and adapt to complex scenarios.

[0091] Shortest - time optimization: Aiming at the shortest time, it meets the user's demand for efficient travel. Combining real - time traffic flow data, dynamically adjusts path planning, avoids congested sections, and further shortens the driving time.

[0092] Strong robustness: Supports dynamic obstacle avoidance and replanning to ensure the safe driving of the vehicle.

[0093] It should be noted that not all steps and modules in the above - mentioned processes and device structures are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted according to needs. The system structure described in the above - mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices may be jointly implemented.

[0094] The above-described embodiments are only preferred embodiments given to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention are all within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A vehicle driving path planning method, characterized in that include: Step 1: Data collection and preprocessing: Collect multi-source data, including GPS data, sensor data and road network information. Preprocess the collected data. Step 2: Perform environmental modeling based on the preprocessed road network information: Construct a dynamic road network model G = (V, E), where V represents the nodes in the road network, E represents the edges between nodes, and the weight ω(e) of the dynamic road network model represents the time cost of passing through the road section. Step 3: Perform a path search: Taking the shortest time as the optimization goal, the improved A* algorithm is used for path search, so that the formula of the heuristic function h(n) is: Where S max is the maximum speed allowed in the road network, (x goal -y goal ) is the target position, (x n -y n ) is the current location, Then the cost function f(n) = g(n) + h(n), where g(n) is the actual time cost from the starting point to node n. During the search process, the weight ω(e) of the dynamic road network model is updated. Step 4: Perform dynamic obstacle avoidance: Dynamic obstacles in the surrounding environment are detected based on sensor data. If an obstacle is detected, the local path is adjusted to avoid the obstacle. If the adjustment of the local path cannot avoid the obstacle, global replanning is triggered and the path search is performed again. Step 5: Output the final path and generate vehicle control instructions based on the final path to control the vehicle.

2. A vehicle driving path planning method according to claim 1, characterized in that The GPS data collected in step 1 includes the vehicle's current location, speed, and direction. The sensor data collected includes the surrounding environment information collected by lidar and cameras. The collected road network information includes road topology, lane information, traffic rules, and real-time traffic flow data.

3. A vehicle driving path planning method according to claim 1 or 2, characterized in that In step 1, the collected data is preprocessed, including: data cleaning, data coordinate alignment, and data time synchronization.

4. A vehicle driving path planning method according to claim 1, characterized in that In step 2, use the formula Calculate ω(e), where L(e) is the length of the road section, S(e) is the average speed of the road section, and T delay (e) is the road section delay time.

5. A vehicle driving path planning device, characterized in that It includes acquisition and preprocessing module, modeling module, search module, obstacle avoidance module and output module. The acquisition and preprocessing module performs data acquisition and preprocessing: Collect multi-source data, including GPS data, sensor data and road network information. Preprocess the collected data. The modeling module performs environmental modeling based on the preprocessed road network information: Construct a dynamic road network model G = (V, E), where V represents the nodes in the road network, E represents the edges between nodes, and the weight ω(e) of the dynamic road network model represents the time cost of passing through the road section. Search modules for path search: Taking the shortest time as the optimization goal, the improved A* algorithm is used for path search, so that the formula of the heuristic function h(n) is: Where S max is the maximum speed allowed in the road network, (x goal -y goal ) is the target position, (x n -y n ) is the current location, Then the cost function f(n) = g(n) + h(n), where g(n) is the actual time cost from the starting point to node n. During the search process, the weight ω(e) of the dynamic road network model is updated. The obstacle avoidance module performs dynamic obstacle avoidance: Dynamic obstacles in the surrounding environment are detected based on sensor data. If an obstacle is detected, the local path is adjusted to avoid the obstacle. If the adjustment of the local path cannot avoid the obstacle, global replanning is triggered and the path search is performed again. The output module outputs the final path and generates vehicle control instructions according to the final path to control the vehicle driving.

6. A vehicle driving path planning device according to claim 5, characterized in that The GPS data collected by the acquisition and preprocessing module includes the vehicle's current location, speed, and direction. The sensor data collected includes the surrounding environment information collected by lidar and cameras. The collected road network information includes road topology, lane information, traffic rules, and real-time traffic flow data.

7. A vehicle driving path planning device according to claim 5, characterized in that collection The collected data is preprocessed with the preprocessing module, including: data cleaning, data coordinate alignment, and data time synchronization.

8. A vehicle driving path planning device according to claim 5, characterized in that The modeling module uses the formula Calculate ω(e), where L(e) is the length of the road section, S(e) is the average speed of the road section, and T delay (e) is the road section delay time.

Images