CN-121977594-A - Unmanned delivery vehicle path planning method and system

CN121977594ACN 121977594 ACN121977594 ACN 121977594ACN-121977594-A

Abstract

The invention provides a method and a system for planning a path of an unmanned delivery vehicle, which relate to the technical field of intelligent logistics, wherein the method comprises the steps of obtaining delivery task point information and environment map data; the method comprises the steps of carrying out joint optimization on a distribution task point access sequence and path nodes through improvement of a sparrow search algorithm to generate an initial global path, carrying out segmentation processing on the initial global path to obtain a plurality of path segments, carrying out local dynamic obstacle avoidance through a dynamic artificial potential field method of real-time fusion sensor data to generate a local correction path, carrying out weight analysis on a plurality of optimization targets through a hierarchical analysis method to construct a multi-target optimization model, embedding the multi-target optimization model into a hybrid optimization framework formed by the improvement of the sparrow search algorithm and the dynamic artificial potential field method, carrying out global and local collaborative optimization on the local correction path to obtain a global optimization path, and verifying the global optimization path through a vehicle dynamics model in a simulation environment to output a final planning path.

Inventors

LI PENG
LIU NIAN
CHEN HAO
Zheng Binshuang

Assignees

杭州中威电子股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251229

Claims (10)

1. A method of unmanned delivery vehicle path planning, comprising: s1, acquiring distribution task point information and environment map data; s2, based on the distribution task point information and the environment map data, performing joint optimization on the distribution task point access sequence and the path nodes by improving a sparrow search algorithm to generate an initial global path; S3, carrying out segmentation processing on the initial global path to obtain a plurality of path segments; S4, based on each path segment, carrying out local dynamic obstacle avoidance by a dynamic artificial potential field method of real-time fusion of sensor data to generate a local correction path; s5, carrying out weight analysis on a plurality of optimization targets comprising path length, safety and energy consumption by using an analytic hierarchy process, and constructing a multi-target optimization model for path planning; s6, embedding the multi-objective optimization model into a hybrid optimization framework formed by the improved sparrow search algorithm and the dynamic artificial potential field method, and carrying out global and local collaborative optimization on the local correction path to obtain a global optimization path; and S7, in the simulation environment, verifying the global optimization path through a vehicle dynamics model, and outputting a final planning path.
2. The unmanned vehicle path planning method according to claim 1, wherein S2 specifically comprises: S201, generating a global skeleton path and a distribution point access sequence through the improved sparrow search algorithm based on the distribution task point information; s202, constructing an environment model containing static barriers and distribution points; S203, generating the initial global path according to the global skeleton path, the distribution point access sequence and the environment model and combining the environment map data.
3. The unmanned vehicle path planning method according to claim 1, wherein the step S3 is specifically that the initial global path is segmented by calculating an optimal delivery sequence to obtain a plurality of path segments.
4. The unmanned vehicle path planning method according to claim 1, wherein S4 specifically comprises: s401, detecting dynamic obstacles and static obstacles by fusing the sensor data in real time, and calculating the movement speed of the dynamic obstacles; s402, respectively calculating a dynamic obstacle repulsive field and a static obstacle repulsive field according to detection results of the dynamic obstacle and the static obstacle; The gain of the dynamic obstacle repulsive field is set to be higher than that of the static obstacle repulsive field, and the strength of the dynamic obstacle repulsive field is enhanced through a speed sensitive factor based on the movement speed of the dynamic obstacle; s403, vector superposition is carried out on the dynamic obstacle repulsive field, the static obstacle repulsive field and the path target gravitational field, so as to obtain total force for path correction; s404, in the vehicle motion model, the path segment is corrected by superposing the total force, and the local correction path is generated.
5. The unmanned vehicle path planning method according to claim 1, wherein S5 specifically comprises: s501, defining an optimization objective function and constraint conditions, wherein the optimization objective function comprises a path length function, a safety function and an energy consumption function; S502, determining the weight of each optimized objective function through the analytic hierarchy process; and S503, carrying out weighted fusion on the weights of the optimization objective functions to form an overall objective function of the multi-objective optimization model, and completing construction of the multi-objective optimization model.
6. The unmanned vehicle path planning method according to claim 5, wherein S502 specifically comprises: S5021, comparing importance of the optimization objective functions in pairs according to expert experience and scene requirements, and constructing a judgment matrix; S5022, calculating to obtain a weight vector by a eigenvalue method based on the judgment matrix; And S5023, carrying out consistency test on the weight vectors, and determining the weight vectors as the weight of each optimized objective function after the consistency test is passed.
7. The unmanned vehicle path planning method according to claim 1, wherein S6 specifically comprises: S601, performing global optimization on the local correction path through the improved sparrow search algorithm on the basis of the multi-objective optimization model at a global optimization layer of the hybrid optimization framework; s602, carrying out dynamic obstacle avoidance correction on a path after global optimization on the basis of the multi-objective optimization model through the dynamic artificial potential field method at a local correction layer of the hybrid optimization framework; S603, ensuring that the curvature of the path after the dynamic obstacle avoidance correction does not exceed a preset maximum curvature through a path smoothing module in the hybrid optimization framework; S604, carrying out collaborative iterative optimization on the outputs of the global optimization layer and the local correction layer through the mixed optimization framework to obtain the global optimization path.
8. The unmanned vehicle path planning method according to claim 1, wherein S7 specifically comprises: S701, constructing a test scene containing dynamic barriers, static barriers and a plurality of distribution nodes in the simulation environment; S702, configuring the vehicle dynamics model based on a monorail bicycle model based on the test scene, and setting key motion parameters of a vehicle; s703, inputting the global optimized path into the vehicle dynamics model to perform path tracking simulation; S704, collecting performance data in the simulation process; And S705, verifying the feasibility of the global optimization path based on the evaluation result of the performance data, and outputting the final planning path after verification.
9. An unmanned delivery vehicle path planning system is characterized by comprising a processor and a memory; The memory stores a program or instructions executable on the processor, which when executed by the processor, implement the steps of the unmanned vehicle path planning method of any one of claims 1 to 8.
10. A readable storage medium having stored thereon a program or instructions which when executed by a processor performs the steps of the unmanned vehicle path planning method of any of claims 1 to 8.

Description

Unmanned delivery vehicle path planning method and system Technical Field The invention relates to the technical field of intelligent logistics, in particular to a method and a system for planning a path of an unmanned delivery vehicle. Background Under the push of intelligent logistics and intelligent city construction, unmanned delivery vehicles have become key vehicles for solving the problem of 'last kilometer' terminal delivery. The advancement and reliability of unmanned delivery vehicle path planning technology directly determine delivery efficiency, operation cost and system safety, and are the core research direction in the technical field of current logistics engineering. Currently, the prior art applied to vehicle path planning mainly includes deterministic algorithms based on graph search and optimization algorithms based on bionic intelligence. A typical solution is to use an improved a-algorithm to perform optimal path search on a global grid map, and combine with an artificial potential field method to perform real-time local obstacle avoidance. However, the optimization algorithm in the prior art ignores the influence of the running environment of the unmanned vehicle on the path planning compared with the running environment of the general traditional vehicle, so that the calculation of the distribution path is lack and the coordination between the mixed traffic flow is caused in the urban high-density environment, the running condition of the unmanned vehicle in the real environment cannot be fully considered in the planning of the distribution path, and the large-scale application of the unmanned vehicle in the complex urban scene is severely restricted. Disclosure of Invention In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a method for planning a path of an unmanned vehicle, which can solve the technical problem that in the prior art, an optimization algorithm ignores the influence of different traveling environments of the unmanned vehicle and general conventional vehicles on the path planning, resulting in the lack of coordination between calculation of the distribution path and mixed traffic flow in a high-density environment of a city, so that the planning of the distribution path cannot fully consider the traveling condition of the unmanned vehicle in a real environment, and severely restricts the large-scale application of the unmanned vehicle in a complex city scene. In a first aspect of an embodiment of the present invention, a method for planning a path of an unmanned delivery vehicle is provided, including: s1, acquiring distribution task point information and environment map data; s2, based on the distribution task point information and the environment map data, performing joint optimization on the distribution task point access sequence and the path nodes by improving a sparrow search algorithm to generate an initial global path; s3, carrying out segmentation processing on the initial global path to obtain a plurality of path segments; S4, based on each path segment, carrying out local dynamic obstacle avoidance by a dynamic artificial potential field method of real-time fusion of sensor data to generate a local correction path; s5, carrying out weight analysis on a plurality of optimization targets comprising path length, safety and energy consumption by using an analytic hierarchy process, and constructing a multi-target optimization model for path planning; s6, embedding a multi-target optimization model into a hybrid optimization framework formed by an improved sparrow search algorithm and a dynamic artificial potential field method, and performing global and local collaborative optimization on the local correction path to obtain a global optimization path; And S7, verifying the global optimization path through a vehicle dynamics model in a simulation environment, and outputting a final planning path. In a second aspect of the embodiment of the invention, an unmanned delivery vehicle path planning system is provided, which comprises a processor and a memory; The memory stores a program or instructions executable on the processor, which when executed by the processor, implement the steps of the unmanned vehicle path planning method of the first aspect. In a third aspect of the embodiments of the present invention, a readable storage medium is provided, on which a program or instructions are stored, which when executed by a processor, implement the steps of the unmanned vehicle path planning method according to the first aspect. The technical scheme provided by the embodiment of the invention has the beneficial effects that at least: In the embodiment of the invention, the multi-objective optimization model based on the analytic hierarchy process is embedded into the hybrid optimization framework formed by the improved sparrow search algorithm and the dynamic artificial potential field method, so that systemati