Heat conduction path topological optimization based underwater glider cruise path planning method

Abstract

A heat conduction path topological optimization based underwater glider cruise path planning method is characterized in that: firstly, a path planning problem is converted into a heat conduction problem, a heat conduction finite element model is established and modeled after the shape of a planning space, an starting point is a heat source, and an end point is a heat outlet; and then heat conduction pipes are arranged with the minimum heat dissipation weak degree as an evaluating standard, the heat source communicates with the heat outlet, at this time, the obtained heat conduction arrangement line is a path of an underwater glider. The method can effectively avoid the situation that the a path planning result falls into a locally optimal solution, can be applied to path planning in a continuous space with a wavy terrain and intricate ocean currents, and can be applied to path planning of unmanned aerial vehicles and all-terrain vehicles; and through the growth type topology optimization method, the method improves the speed and the quality of path planning; finite element grids are processed through a heat rigidity equivalent method, two ends of the heat conduction pies do not need to be fixed on nodes established by a man like the conventional topology optimization method, and the obtained result is close to the optimal solution.

Description

technical field [0001] The invention belongs to the technical field of robot path planning, and in particular relates to an underwater glider cruise path planning method based on thermal conduction path topology optimization. technical background [0002] As a kind of robot, the common path planning methods of underwater glider can be divided into four categories: template matching method, artificial potential field method, map construction method and artificial intelligence planning method; among them, the template matching method relies too much on existing Some templates have poor adaptability to unfamiliar environments; artificial potential field methods tend to fall into local optimal solutions, and are difficult to adapt to complex terrain with scattered heights; map construction methods must rely on fine grid division to obtain satisfactory results. The amount is large, and real-time performance cannot be guaranteed, especially in three-dimensional space. The artifici...

AI Technical Summary

Problems solved by technology

[0002] As a kind of robot, the common path planning methods of underwater glider can be divided into four categories: template matching method, artificial potential field method, map construction method and artificial intelligence planning method; among them, the template matching method relies too much on existing Some templates have poor adaptability to unfamiliar environments; artificial potential field methods tend to fall into local optimal solutions, and are difficult to adapt to complex terrain with scattered heights; map construction methods must rely on fine grid division to obtain satisfactory results. The amount is large, and real-time performance cannot be guaranteed, especially in three-dimensional space. The artificial intelligence planning method requires a long learning process, and the real-time performance is poor, especially in unfamiliar environments.

Images