Research Of Mobile Robot Path Planning In Complex Environment Based On Grey Wolf Optimization

In the natural and social environment that people depend on for survival,natural or man-made disasters such as fire,flood,earthquake,landslide,explosion,and war exist objectively,while threatening people's lives and property directly,also bringing harsh and complex rescue environmental conditions,causing huge difficulties to rescue operations.With the development of science and technology,the manufacturing level,execution ability,information interaction level and intelligence of robots are getting better and better.They have been successfully used in household,industry,public security,military and other fields,they have obvious advantages and special value especially in harsh environmental conditions.It has been a hot spot for rescue departments,scientific research institutions and scientific enterprises to utilize robots,the automatic method,to facilitate fast search and precise aid by rescuers and to guarantee the safety of people.Rescue after disaster encounters diverse environments and complex tasks.The Swarm-robot system bears prominent application advantages in rescue with its strong independence and fine adaptability.Under such circumstances,the key point of rescue rapidity and efficiency lies in route planning and task allocation strategies of swarm-robot,which is also one of the puzzles in swarm-robot research.Aiming at problems above,this paper takes self-organizing search strategies of swarm-robot system as object of study and conducts research into route planning and formation of self-organization formative algorithm of swarm-robot in two-dimension environment.First of all,modified artificial moment method is adopted to study and analyze the obstacle-avoiding issues in robot route planning and coordinated control of swarm-robot is designed to form simple formation formative algorithm.Based on such operations,self—organizing cooperate search algorithm of swarm-robot is proposed on the ground of grey wolf optimization algorithm;basic grey wolf optimization is introduced,the convergence factor is improved non-linearly,and the advantage of the modified version is tested with function testing method.Subsequently,combining artificial moment method and grey wolf optimization,multi-target search and rescue strategies of swarm-robot are designed,including search of the environment at initial stage by robots in the system,location and switch of rescue target on approach and robots' internal collision-avoiding methods on the move.The final step is to conduct H-pattern,multi-target search and rescue simulation of rectangle-featured objects and multi-target search and rescue simulation in environment with obstacles based on MATLAB simulation platform,whose result testify the algorithm's advantages such as accuracy,reliability,high-efficiency and low computation.With the research mentioned,artificial moment method and grey wolf optimization,a set of swarm-robot self-organization coordinated search and rescue strategies are formulated.Through modifying artificial moment controller algorithm,robots are able to avoid obstacles efficiently and can avoid falling into local minimum value and pass through narrow channels.With the target nodes switch control of grey wolf optimization algorithm,each robot can be assigned with the most appropriate rescue target,thus improving the rescue efficiency.The strategy effectively improves the obstacle-avoidance and coordinated search and rescue capacity of swarm-robot in complicated environment and optimizes the efficiency of comprehensive rescue.