Rearch Of Path Planning And Task Allocation In RoboCup Rescue Simulation System

With the improvement of social productivity, human civilization will also continue to move forward.The urbanization process of each country and region in the world is accelerating.high density of population.buildings and roads are striking features of urbanization. However.the serious consequence of frequent natural disasters in recent years,let people began to pay attention to disaster relief activities.Because of the risk and complexity of the disaster rescue,agent become to the most ideal candidates for rescue.Rescue plan and theoretical research after the disaster should have a rescue simulation platform.RoboCup Rescue Simulation System is the simulation of the real city disaster scenarios by using computer technology.Based on the simulation of agent rescue activities,we want to achieve the maximum possible of fire fighting and rescue people, the smallest possible to reduce the losses caused by disasters.Outstanding rescue strategy and method in RoboCup Rescue Simulation platform will provide the technical support for people in the actual rescue activities.So the study is of great practical significance.In the RoboCup Rescue Simulation System, fire protection has an important influence on the whole system of the rescue effect.it is not difficult to make sure of it from the evaluation standards.Fire mission is completed by the fire agent, and the effective control of the fire will often play a decisive role to win over more rescue time for other rescue agents to complete the mission of rescue people.So, the outstanding fire agent path planning and task allocation method is particularly important for the whole system. Based on the RoboCup Rescue Simulation System, the main research is as follows:1) Fire Agent task allocation methodIn case of the global map information with center agent, the method based on the distance task allocation is simple and reliable.Because of the existence of the central node, we can put the task set into a list, the fire agent into another list.Then calculate the distance between the task and the fire agent, we can get the most reasonable task allocation. In case of the local map information without center agent,we can use contract net protocol completes the fire Agent task allocation, through the "invite bids-bidding-win the bidding-contract" four steps to achieve the application of contract net protocol task allocation in Robocup Rescue Simulation System.Finally, proved the methods are effectively and practicable.2) Fire Agent path planningAccording to the task allocation method, we just get the most reasonable fire agent task allocation.In another word, we get the starting point and the target point of each fire agent. There are many kinds of path for fire Agent from the starting point to the target point, and the shortest road calculated by the theory cannot reach because of the blockade in the road. Therefore, how to plan a path to make the system cost and reach time minimum in the dynamic environment is extremely important. Ant Colony Algorithm is a global search method with the feature of heuristic and distributed, especially suitable for the application in Robocup Rescue Simulation System and obtain the optimal path planning for fire agent. Finally,through the experiment using this method to achieve the most reasonable path from the starting point to the target point.At the end of this dissertation, the main research is summarized. It makes out the points of innovation and main research results, also give the shortage in the dissertation and the problem need to further research.