Research On Low-energy Path Planning And Priority Decoupling Of Robot

The classical ant colony algorithm solves the paths of mobile robots in complex unstructured rough environments with a single optimization objective and constraint,and the requirement of sustainable robot operation cannot be satisfied and is not applicable to highly dynamic multi-robot environments.In this paper,we consider multiple kinematic energy consumption factors of mobile robots in complex unstructured rough environments to explore single-robot low-energy path planning under improved ant colony algorithm optimization,and combine priority rules to solve multiple collision and conflict problems among robots to achieve multi-robot low-energy conflict-free path planning,and the main research contents are as follows:(1)A two-layer raster map model is developed.A two-layer raster map model is built based on the traditional raster map model considering the ground friction coefficient information in the complex unstructured rough environment.The model converts the obstacle information and ground friction coefficient information in the map modeling into constraint factors by combining the robot’s motion capability.(2)Considering various factors to improve the traditional ant colony algorithm.In order to reduce the energy loss during robot movement,this paper improves the heuristic function by considering the energy consumption factors(path length,path smoothness,ground friction coefficient,and energy consumption of ants in the process of grid transfer)during the path search,and improves the heuristic function by considering various energy consumptions during robot movement(the energy consumption of robot starting/braking and turning process,the energy consumption of robot overcoming friction The pheromone update strategy is improved by considering multiple energy consumptions during robot motion(energy consumptions during robot start/braking and turning,energy consumptions during robot overcoming friction,energy consumptions during energy conversion by robot’s own hardware devices),so as to guide the subsequent ant colony to evolve in the direction of optimal total energy consumption to achieve a low-energy path for the robot.Finally,the effectiveness of the multi-factor improved ant colony algorithm in reducing energy consumption was tested by three neighborhood search methods,especially in the four-neighborhood search method,the multi-factor improved ant colony algorithm outperformed other methods by 11.32%,8.39%,13.16%,19.91%,and 26.30%,respectively.(3)Energy consumption priority dissipation of multi-robot collision conflicts.For the problem of multiple motion conflicts(node conflict,alignment conflict,occupancy conflict,blocking conflict,unresolved conflict 1,unresolved conflict 2)caused by high coupling of multi-robot paths,a path conflict deconfliction strategy based on energy consumption ascending priority rule is used to initially deconflict robot path conflicts;then a priority conflict deconfliction strategy is used to deeply decouple the conflicts that still exist between robots.Finally,the energy consumption priority rule in the obstacle-free environment verifies that the energy consumption ascending priority rule is 13.52% and 7.80% better than the conventional priority rule and energy consumption descending priority rule,respectively,in reducing the path energy consumption;the multi-robot large-scale conflict decoupling experiment verifies the effectiveness of the above conflict decoupling strategy in planning low-energy conflict-free paths for multiple robots.