Unmanned aerial vehicle adaptive repulsion coefficient path planning method based on deep learning

Abstract

The invention discloses an unmanned aerial vehicle adaptive repulsion coefficient path planning method based on deep learning. The method comprises the following steps: firstly, finding out a most suitable repulsion coefficient sample set in a specific environment through a fusion genetic algorithm and an artificial potential field method; secondly, training a residual neural network by using the sample set; and finally, a repulsive force coefficient adapting to the environment is calculated through the residual neural network, and path planning is performed by using an artificial potential field method. The problems that a traditional artificial potential field method cannot adjust the repulsive force coefficient according to specific environment information, and an existing improvement method cannot give consideration to the planning effect and the planning duration while adaptively adjusting the repulsive force coefficient are solved. Simulation experiments show that the method has excellent performance in the aspects of planning effect and planning duration, and can well meet the self-adaption requirement and the rapid planning requirement for the current environment in practical application.

Description

technical field [0001] The invention relates to the field of UAV path planning, in particular to an UAV adaptive repulsion coefficient path planning method based on deep learning. Background technique [0002] Compared with manned aerial vehicles, unmanned aerial vehicles (UAVs) have the advantages of small size, low cost, flexible use, and no casualties. With the continuous development of aviation technology and automation technology, UAVs have been widely used in military, agriculture, transportation, public management and other fields. For example, when natural disasters such as mudslides and landslides occur, drones can be used to understand the disaster situation safely and quickly; when haze weather occurs, drones carrying catalysts can be used to spray in the air to clear the haze; In control, UAVs can immediately detect emergencies and implement emergency rescue; in military, UAVs have greatly cooperated in the completion of tasks such as reconnaissance, target acqu...

AI Technical Summary

Problems solved by technology

[0004] However, the repulsion coefficients in these improved methods based on the artificial potential field method are all set based on experience, and are constant, and cannot adapt to changes in the environment, resulting in problems such as unreachable targets and local minimum values ​​in the planned path. Issues such as pitfalls and excessive path lengths

However, the existing improved methods cannot adaptively adjust the repulsion coefficient while taking into account the planning effect and planning time.

Images