Study on the path planning of the internal spiral complete traversal for glass-curtain wall-cleaning robots
In high-rise buildings and complex environments,it is difficult to realize the omnidirectional cleaning of glass-curtain wall-cleaning robots.To address this problem,this paper presents a path-planning method for internal spiral complete traversal based on the improved ant colony algorithm.First,the environment raster map was con-structed based on the fusion of vision sensor recognition and positioning results.Subsequently,for the robot's omni-directional traversal with the nondead angel problem,cleaning path planning was realized by the internal spiral complete traversal algorithm with the objectives of maximum coverage area,minimum repetition path,and maximum safety.On this basis,the optimal route to escape the dead zone was planned using the improved ant colony algo-rithm with the objectives of fastest planning,least inflection points,and fastest convergence.The improved direc-tion was based on the robot's four-direction and four-field movement,and the cost function of the A∗ algorithm was introduced in the heuristic function,while the idea of a penalty function was introduced in the pheromone update.Finally,the complete traversal of the cleaning path was completed by combining the two algorithms.The simulation results revealed that after the robot passed four times of planning on a known map,the cleaning coverage was 100%,and the travel repetition rate was 3.15%,realizing complete traversal.
glass-curtain wallrobotpath planningant colony algorithminternal spiral coverage algorithmA∗algorithmheuristic functionpenalty function
国家科技期刊平台
NSTL
万方数据
