Research On Multiple Agent Rescue Simulation System

Over the last few years, a Multi-Agent System(MAS) has been the important branch in the field of distributed artificial intelligence. This thesis is based on the study of the RoboCupRescue simulation system, due to the dynamics complex rescue environment, the multi-agent system's path planning, task allocation and task cooperation in dynamic environment are introduced. The main research contents and contributions are listed as follows.Robot rescue Simulation System (RoboCupRescue Simulation System, RCRSS) is a typical heterogeneous multi-agent System. It is also a real-time distributed Simulation System. The first part of the thesis summaries the essential elements, the simulation process and the controlling method in the RoboCupRescue simulation system to lay a solid foundation for the platform's application.A path search method called improved ant algorithm is presented to solve the shortest path search problem of the rescue robot in the city with dynamic topology after earthquake. The improved ant algorithm is used as heuristic function together with pheromone intensity to guide the ant's searching and make ants chose the nodes which is closer to the target node, speeding up the convergence rates of the ant colony and enabling ants converge to the shortest path as possible as they can, avoiding converging to local optimum.A Multi-Agent task allocation algorithm based on auction is proposed to solve the task coordination problem of the homogeneous agents. This algorithm mainly includes the task allocation scheme and the greedy strategy. By introducing the auction algorithm a nearly optimized task allocation scheme can be inferred after a comprehensively evaluation to the benefit and cost of the task. AND take real-time dynamic adjustment to the task circumstance of agents by greedy strategy until to the current optimal state.To overcome the problems such as the task interdependence, the Q-Learning algorithm is used in the Multi-Agent task cooperation. Strengthen the positive effect of multi-agent rescue to solve the multi-agent task interdependence and constraint relationship.The proposed algorithms of the dissertation have been applied to the robot rescue simulation team of Harbin Engineering University, which won the third place in 2010 China Robot Contest.And that verified the efficiency of the proposed algorithms.