CN-121996350-A - Multi-unmanned aerial vehicle and multi-sensor intelligent task matching system

CN121996350ACN 121996350 ACN121996350 ACN 121996350ACN-121996350-A

Abstract

The application discloses a multi-unmanned aerial vehicle and multi-sensor intelligent task matching system and method, and relates to the technical field of unmanned aerial vehicles. The method comprises the steps of obtaining task information of tasks to be distributed and real-time state information of a plurality of unmanned aerial vehicles, constructing a space-time K-D tree index based on four-dimensional space-time coordinates contained in the unmanned aerial vehicles, performing range query in the space-time K-D tree index for each task to be distributed based on space-time attributes defined by the task information of the task to screen candidate unmanned aerial vehicle sets meeting conditions so as to generate sparse candidate matching pair sets, calculating matching benefits only for matching pairs in the sparse candidate matching pair sets, and solving an optimal task distribution scheme based on calculation results. According to the application, the space-time index is constructed to perform pre-pruning, so that the solving scale and the calculating complexity of the matching problem are obviously reduced, and the real-time performance and the efficiency of task matching of the large-scale unmanned aerial vehicle cluster in a dynamic environment are improved.

Inventors

YAN SHAOBIN
YAN FEIFEI
XUE YAOHUI

Assignees

上海博昂电气有限公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (10)

1. A multi-unmanned aerial vehicle and multi-sensor intelligent task matching method is characterized by comprising the following steps: Acquiring task information corresponding to at least one task to be allocated and real-time state information of a plurality of unmanned aerial vehicles; Constructing an index structure based on real-time state information of each unmanned aerial vehicle in the unmanned aerial vehicles, wherein the index structure is used for representing multi-dimensional space-time distribution of the unmanned aerial vehicles; Inquiring in the index structure based on any target task in the tasks to be distributed to determine one or more candidate unmanned aerial vehicles meeting the space-time inquiry range in the unmanned aerial vehicles, and generating candidate matching pairs between the target task and the candidate unmanned aerial vehicles; And determining an optimal task allocation scheme between the task to be allocated and the unmanned aerial vehicle based on the candidate matching pair.
2. The method of claim 1, wherein constructing an index structure based on real-time status information of each of the drones includes: Extracting four-dimensional space-time coordinate points representing the three-dimensional space position and the residual endurance time of each unmanned aerial vehicle from the real-time state information of each unmanned aerial vehicle; and constructing a space-time K-D tree as the index structure based on all four-dimensional space-time coordinate points corresponding to the unmanned aerial vehicle.
3. The method of claim 2, wherein querying the index structure based on any target task of the tasks to be assigned comprises: Extracting a task position, a task influence range, task starting time and task ending time from the task information of the target task; Determining a three-dimensional space query boundary based on the task position and the task influence range; Determining a one-dimensional time query boundary based on the task start time and the task end time; And determining a space-time query range by the three-dimensional space query boundary and the one-dimensional time query boundary.
4. The method of claim 1, wherein the determining an optimal task allocation scheme between a task to be allocated and the drone based on the candidate matching pair comprises: Calculating a matching profit value based on task information of a corresponding task and real-time state information of the unmanned aerial vehicle for each candidate matching pair of the candidate matching pairs; constructing a sparse gain matrix based on all candidate matching pairs and corresponding matching gain values thereof; and determining the optimal task allocation scheme based on the sparse revenue matrix.
5. The method according to claim 1, wherein the method further comprises: Continuously monitoring real-time state information and external environment information of the unmanned aerial vehicle; and when the occurrence of a preset dynamic scheduling triggering event is monitored, the optimal task allocation scheme is reconfirmed.
6. A multi-unmanned aerial vehicle, multi-sensor intelligent task matching system, comprising: The information acquisition module is used for acquiring task information corresponding to at least one task to be allocated and real-time state information of a plurality of unmanned aerial vehicles; the index construction module is used for constructing an index structure based on real-time state information of each unmanned aerial vehicle in the unmanned aerial vehicles, and the index structure is used for representing multi-dimensional space-time distribution of the unmanned aerial vehicles; The candidate screening module is used for inquiring in the index structure based on any target task in the tasks to be distributed so as to determine one or more candidate unmanned aerial vehicles meeting the space-time inquiry range in the unmanned aerial vehicle, and generating candidate matching pairs between the target task and the candidate unmanned aerial vehicles; and the task allocation module is used for determining an optimal task allocation scheme between the task to be allocated and the unmanned aerial vehicle based on the candidate matching pair.
7. The system of claim 6, wherein the index building module is to: The constructing the index structure based on the real-time state information of each unmanned aerial vehicle comprises the following steps: Extracting four-dimensional space-time coordinate points representing the three-dimensional space position and the residual endurance time of each unmanned aerial vehicle from the real-time state information of each unmanned aerial vehicle; and constructing a space-time K-D tree as the index structure based on all four-dimensional space-time coordinate points corresponding to the unmanned aerial vehicle.
8. The system of claim 6, wherein the task allocation module is configured to: Calculating a matching profit value based on task information of a corresponding task and real-time state information of the unmanned aerial vehicle for each candidate matching pair of the candidate matching pairs; constructing a sparse gain matrix based on all candidate matching pairs and corresponding matching gain values thereof; and determining the optimal task allocation scheme based on the sparse revenue matrix.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program, wherein the computer program is arranged to execute the method of any of the claims 1 to 5 when run.
10. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of any of the claims 1 to 5.

Description

Multi-unmanned aerial vehicle and multi-sensor intelligent task matching system Technical Field The application relates to the technical field of unmanned aerial vehicles, in particular to a multi-unmanned aerial vehicle and multi-sensor intelligent task matching system and method. Background Along with the rapid development of unmanned aerial vehicle technology, the application of the multi-unmanned aerial vehicle collaborative operation system in the fields of logistics, inspection, security, mapping and the like is increasingly wide. In a complex operation scene, how to distribute a large number of dynamic tasks to a plurality of heterogeneous unmanned aerial vehicles in real time and efficiently is a core technical problem for ensuring the operation efficiency and stability of the whole system. The existing task allocation method mostly adopts the steps of constructing a global cost or benefit matrix, and solving by combining a Hungary algorithm, an auction algorithm and the like. However, when the number of tasks and unmanned aerial vehicles increases sharply, the computational complexity of building and solving the global matrix increases exponentially, resulting in high decision delay, and difficulty in adapting to dynamic environments with sudden task and rapid unmanned aerial vehicle state change, thereby constituting a core bottleneck of the current technology. Disclosure of Invention The application aims to provide a multi-unmanned aerial vehicle and multi-sensor intelligent task matching system and method, and aims to solve the technical problems of high complexity and poor real-time performance of large-scale task matching calculation in the prior art. In a first aspect, the application provides a multi-unmanned aerial vehicle and multi-sensor intelligent task matching method, which comprises the steps of obtaining task information corresponding to at least one task to be distributed and real-time state information of a plurality of unmanned aerial vehicles; Constructing an index structure based on real-time state information of each unmanned aerial vehicle in the unmanned aerial vehicles, wherein the index structure is used for representing multi-dimensional space-time distribution of the unmanned aerial vehicles; Inquiring in the index structure based on any target task in the tasks to be distributed to determine one or more candidate unmanned aerial vehicles meeting the space-time inquiry range in the unmanned aerial vehicles, and generating candidate matching pairs between the target task and the candidate unmanned aerial vehicles; And determining an optimal task allocation scheme between the task to be allocated and the unmanned aerial vehicle based on the candidate matching pair. Optionally, the constructing the index structure based on the real-time status information of each of the unmanned aerial vehicles includes: Extracting four-dimensional space-time coordinate points representing the three-dimensional space position and the residual endurance time of each unmanned aerial vehicle from the real-time state information of each unmanned aerial vehicle; and constructing a space-time K-D tree as the index structure based on all four-dimensional space-time coordinate points corresponding to the unmanned aerial vehicle. Optionally, the querying in the index structure based on any target task in the tasks to be distributed includes: Extracting a task position, a task influence range, task starting time and task ending time from the task information of the target task; Determining a three-dimensional space query boundary based on the task position and the task influence range; Determining a one-dimensional time query boundary based on the task start time and the task end time; And determining a space-time query range by the three-dimensional space query boundary and the one-dimensional time query boundary. Optionally, the determining, based on the candidate matching pair, an optimal task allocation scheme between the task to be allocated and the unmanned aerial vehicle includes: Calculating a matching profit value based on task information of a corresponding task and real-time state information of the unmanned aerial vehicle for each candidate matching pair of the candidate matching pairs; constructing a sparse gain matrix based on all candidate matching pairs and corresponding matching gain values thereof; and determining the optimal task allocation scheme based on the sparse revenue matrix. Optionally, the method further comprises: Continuously monitoring real-time state information and external environment information of the unmanned aerial vehicle; when the occurrence of a preset dynamic scheduling trigger event is monitored, the optimal task allocation scheme is reconfirmed In a second aspect, the present application provides a multi-unmanned aerial vehicle, multi-sensor intelligent task matching system, the system comprising: The information acquisition module is used for acquiring task