CN-122022624-A - Unmanned aerial vehicle logistics distribution path planning method, device and equipment

CN122022624ACN 122022624 ACN122022624 ACN 122022624ACN-122022624-A

Abstract

The application discloses a method, a device and equipment for planning logistics distribution paths of unmanned aerial vehicles, wherein the method comprises the steps of constructing a corresponding task model based on a current distribution task when the current distribution task is received, constructing a primary population based on an environment sub-model, a multi-objective cost function and task constraint conditions, carrying out iterative optimization on the primary population by adopting a preset genetic algorithm according to an optimization target of a minimized multi-objective cost function to generate an initial global path planning scheme, splitting the initial global path planning scheme according to unmanned aerial vehicle continuous information, and determining distribution batches and corresponding batch distribution planning paths of all unmanned aerial vehicles. Therefore, the overall efficiency, the resource utilization rate and the dynamic adaptability of the distribution system are remarkably improved through multi-objective collaborative optimization while the order differentiation aging requirement and the unmanned aerial vehicle hardware constraint are met.

Inventors

LI JING
LI YANG
CHEN MING
TAO BAOQUAN
PU XIANKUN

Assignees

湖北文理学院

Dates

Publication Date: 20260512
Application Date: 20251231

Claims (10)

1. A method for planning a logistics distribution path of an unmanned aerial vehicle, the method comprising: When a current distribution task is received, constructing a corresponding task model based on the current distribution task, wherein the task model comprises an environment sub-model, a multi-objective cost function and task constraint conditions; Constructing a primary population based on the environment sub-model and the task constraint condition, and performing iterative optimization on the primary population by adopting a preset genetic algorithm according to an optimization target for minimizing the multi-objective cost function to generate an initial global path planning scheme; Splitting the initial global path planning scheme according to the unmanned aerial vehicle endurance information, and determining the distribution batch and the corresponding batch distribution planning path of each unmanned aerial vehicle.
2. The method of claim 1, wherein upon receiving a current delivery task, the step of constructing a corresponding task model based on the current delivery task comprises: When a current distribution task is received, determining corresponding task area geographic information, and constructing an environment sub-model based on the task area geographic information; determining a cost element associated with the current delivery task, and setting a multi-objective cost function according to the cost element; Determining task constraint conditions according to preset configuration demand information, wherein the task constraint conditions at least comprise order constraint and area constraint; And constructing a task model according to the environment sub-model, the multi-objective cost function and the task constraint condition.
3. The method of claim 2, wherein the step of determining a cost element associated with the current delivery task and setting a multi-objective cost function based on the cost element comprises: Determining cost elements associated with the current delivery task, wherein the cost elements comprise order timeliness cost, unmanned aerial vehicle number cost, total path distance cost and collaboration efficiency cost; determining a task scene to which the current distribution task belongs, and setting a weight coefficient corresponding to the cost element according to the task scene; And constructing the multi-objective cost function according to the cost elements and the corresponding weight coefficients.
4. The method of claim 1, wherein the step of constructing a first generation population based on the environmental submodel and the task constraints comprises: Acquiring set algorithm parameter configuration information, wherein the algorithm parameter configuration information comprises population scale, maximum iteration times, crossover probability and variation probability; Constructing an initial code by adopting a segmentation coding mode based on the environment sub-model according to the algorithm parameter configuration information, wherein the initial code comprises a path segment and a breakpoint segment; Performing validity check on the initial code based on the task constraint condition; upon passing the validity check, the initial code is determined to be a primary population.
5. The method of claim 4, wherein the step of iteratively optimizing the primary population with a preset genetic algorithm according to an optimization objective that minimizes the multi-objective cost function to generate an initial global path planning scheme comprises: Calculating the fitness of each individual in the primary population based on the multi-objective cost function; Screening parent individuals in the primary population according to the fitness; performing cross operation on the parent individuals according to the cross probability, wherein the cross operation at least acts on the path section and/or the breakpoint section; Performing mutation operation on the child individuals generated by the cross operation according to the mutation probability to obtain the child individuals generated by the mutation operation to form a current generation population, wherein the mutation operation is exchange mutation operation, inversion mutation operation and/or disorder mutation operation acting on the path segment; Selecting the child individuals with the fitness in the preset order in the current generation population as new parent individuals, and returning to the step of performing cross operation on the parent individuals according to the cross probability; And determining an initial global path planning scheme according to the current generation population corresponding to the maximum iteration times until the current iteration times reach the maximum iteration times.
6. The method of claim 1, wherein the initial global path planning scheme is a mission-wide path planning scheme corresponding to each drone; Splitting the initial global path planning scheme according to unmanned aerial vehicle duration information, and determining the distribution batch and the corresponding batch distribution planning path of each unmanned aerial vehicle, wherein the method comprises the following steps: Determining the longest single endurance distance of each unmanned aerial vehicle according to the unmanned aerial vehicle endurance information; and acquiring the total distance of all tasks corresponding to each all task path planning scheme, and splitting each all task path planning scheme by combining each longest single cruising distance and each total distance of all tasks to determine the distribution batch and the corresponding batch distribution planning path of each unmanned plane.
7. The method of claim 6, wherein the step of splitting each of the mission-wide path planning schemes in combination with each of the longest single cruising distance and each of the mission-wide total distances to determine a distribution lot and a corresponding lot distribution planning path for each of the unmanned aerial vehicles comprises: Determining available power exchange station position information according to the full-mission path planning scheme; And splitting each full-mission path planning scheme according to the available power exchange station position information, each longest single cruising distance and each full-mission total distance, and determining the distribution batch and the corresponding batch distribution planning path of each unmanned aerial vehicle.
8. The method of claim 1, wherein after the step of determining the distribution lot for each drone and the corresponding lot distribution planning path, further comprising: When a new delivery task is received, determining the execution condition of the current delivery task; Determining a next current delivery task according to the execution condition of the current delivery task and the newly added delivery task, and returning to the step of constructing a corresponding task model based on the current delivery task; And adjusting each batch of distribution planning paths according to the initial global path planning scheme corresponding to the next current distribution task until the initial global path planning scheme corresponding to the next current distribution task is generated, so as to obtain adjusted batch of distribution planning paths.
9. An unmanned aerial vehicle logistics distribution path planning apparatus, the apparatus comprising: The task initialization module is used for constructing a corresponding task model based on a current distribution task when the current distribution task is received, wherein the task model comprises an environment sub-model, a multi-objective cost function and task constraint conditions; The path planning module is used for constructing a first generation population based on the environment sub-model and the task constraint condition, and carrying out iterative optimization on the first generation population by adopting a preset genetic algorithm according to an optimization target for minimizing the multi-objective cost function to generate an initial global path planning scheme; And the batch distribution module is used for splitting the initial global path planning scheme according to the unmanned aerial vehicle continuous voyage information and determining the distribution batch and the corresponding batch distribution planning path of each unmanned aerial vehicle.
10. An unmanned aerial vehicle logistics distribution path planning apparatus comprising a memory, a processor and an unmanned aerial vehicle logistics distribution path planning program stored on the memory and executable on the processor, the unmanned aerial vehicle logistics distribution path planning program configured to implement the steps of the unmanned aerial vehicle logistics distribution path planning method of any one of claims 1 to 8.

Description

Unmanned aerial vehicle logistics distribution path planning method, device and equipment Technical Field The present application relates to the field of path planning technologies, and in particular, to a method, an apparatus, and a device for planning a logistics distribution path of an unmanned aerial vehicle. Background In recent years, a low-altitude logistics network is rapidly expanded, and a plurality of unmanned aerial vehicles in the same space for parallel execution of delivery tasks become a normalized running mode. The order types are subdivided with the order types, so that emergency materials sensitive to time efficiency are contained, community retail goods which can be replenished according to a fixed period are also contained, and the running constraint is further increased in complex airspace environments of cities and suburbs. On the premise of guaranteeing the safety of an airspace, different types of orders are reasonably distributed to a multi-machine formation and high-efficiency flight is completed, and the method is a basic requirement in the current unmanned plane logistics path planning field. In the prior art, path planning is generally regarded as a single-objective optimization problem of the shortest path, and then, a route meeting the course constraint can be obtained by adopting clustering and segmentation solution or classical evolutionary algorithm for order distribution, however, in an actual logistics scene, the order always has different ageing requirements and distribution modes, and the single optimization objective and indiscriminate scheduling strategy adopted in the prior art are difficult to realize efficient cooperation of the whole resource while meeting the needs of diversified orders. Disclosure of Invention The application mainly aims to provide a method, a device and equipment for planning a logistics distribution path of an unmanned aerial vehicle, and aims to solve the technical problem that the conventional unmanned aerial vehicle path planning mode adopting a single optimization target is difficult to meet the distribution requirements of diversified orders. In order to achieve the above purpose, the present application provides an unmanned aerial vehicle logistics distribution path planning method, which includes: When a current distribution task is received, constructing a corresponding task model based on the current distribution task, wherein the task model comprises an environment sub-model, a multi-objective cost function and task constraint conditions; Constructing a primary population based on the environment sub-model and the task constraint condition, and performing iterative optimization on the primary population by adopting a preset genetic algorithm according to an optimization target for minimizing the multi-objective cost function to generate an initial global path planning scheme; Splitting the initial global path planning scheme according to the unmanned aerial vehicle endurance information, and determining the distribution batch and the corresponding batch distribution planning path of each unmanned aerial vehicle. In one embodiment, the step of constructing a corresponding task model based on the current delivery task when the current delivery task is received includes: When a current distribution task is received, determining corresponding task area geographic information, and constructing an environment sub-model based on the task area geographic information; determining a cost element associated with the current delivery task, and setting a multi-objective cost function according to the cost element; Determining task constraint conditions according to preset configuration demand information, wherein the task constraint conditions at least comprise order constraint and area constraint; And constructing a task model according to the environment sub-model, the multi-objective cost function and the task constraint condition. In one embodiment, the step of determining the cost element associated with the current delivery task and setting a multi-objective cost function according to the cost element includes: Determining cost elements associated with the current delivery task, wherein the cost elements comprise order timeliness cost, unmanned aerial vehicle number cost, total path distance cost and collaboration efficiency cost; determining a task scene to which the current distribution task belongs, and setting a weight coefficient corresponding to the cost element according to the task scene; And constructing the multi-objective cost function according to the cost elements and the corresponding weight coefficients. In one embodiment, the step of constructing a first generation population based on the environmental sub-model and the task constraints comprises: Acquiring set algorithm parameter configuration information, wherein the algorithm parameter configuration information comprises population scale, maximum iteration times, crossov