Path Planning For Mobile Robot And Rescue Based On Improved Ant Colony Optimization Algorithm

In recent years, destructive disasters, such as fire, earthquake, city disaster etc. have caused urban people paid close attention. It is difficult for human being to carrying out rescue in this narrow space environment. USA, Japan, China and other countries have combined robot technology, the disaster technology, and multiparty rescue technology and science knowledge in order to develop special robot for disaster urban search and rescue. The disaster scene is a complex, dangerous unknown environment. It is a serve challenge of how to search for and rescue survivors in this kind of environment quickly. The related technical problems about the search and rescue robot path planning are studied in the disaster environment in this paper. Robot path planning is an important research subject in the robot search and rescue fields. According to certain performance index, it should look for the optimal or subprime path from the starting point to target in this complex environment. Robot cannot bump obstacles object during walking.In order to transform the physical map information of search environment into the abstract space which computers can process, this paper study the construction of the environment space model, and comparatively analyze the mature building method about the environment model at present. And then we determine the grid as a method for constructing environment model. This paper discusses the connection between the mathematical matrix for environment and object-oriented raster image, and gives the MATLAB code implementation. Based on the superiority of grids, the paper leads to the graph theory, and uses the adjacent matrix to realize the distance between the grids. Finally, it presents relative strategies for some possible isolated grids.Path optimization algorithm is the most important factor in the path planning. This paper discusses the relevant technology of the ant colony algorithm theory, and analyzes the flaws about the basic ant colony algorithm to solve path problems and suggests some improvement strategies, such as filling and target attracting. The rolling window strategy for the information unpredictability in disaster rescue environment is researched, and the local environmental information is got within the scope of window with rolling window. Local objectives are got in the scope of window through the heuristic search, and it will transform the global path optimization problem into multiple paragraphs local path optimization problem. Finally, the strategies of rolling window and the improved ant colony algorithm are combined to get the realization process and solving steps for the local path.In order to discuss the key technology of robot path planning and the realization of its algorithm, this paper finally develops a simulation experimental platform. The performance evaluation indexes of the improved ant colony algorithm is better than the basic algorithms, so it can complete the disaster path planning problem better combining with the rolling window control. Finally, the feasibility and rationality of improved algorithm is verified, which provides new ideas for robot path planning in the real disaster rescue environment.