Research On Multi-Robot-Multi-Task Coordination Method Based On The Integration Of Intelligent Agent And Petri Net

With the development of population aging,China and even the world will face enormous pressure caused by the growing long-term care needs of the elderly in the future,especially in places like nursing homes,which require a lot of human resources to complete the care of the elderly.With the development of robotics,some service robots have been able to help people care for the elderly to a certain extent.It can be speculated that with the intensification of aging and the relative decrease of labor force,using service robots to care for the daily life of the elderly will become an inevitable trend.Facing a certain number of service robots with different functions,how to make them work together to complete daily care tasks and ensure the satisfaction of the elderly in the process has become a problem to be solved.In order to solve this problem,this thesis proposes a multi-robot-multi-task coordination framework applied to the context of nursing home.In this thesis,firstly,the Scalable Timed-Colored Petri Net(STdCPN)is proposed by combining the Timed Petri Net and the Colored Petri Net,and the process of robots caring for the elderly in the nursing home is modeled by STdCPN,so as to simulate the basic operation of the nursing home,which is the basis of this coordination framework.Secondly,by considering the factors that will affect the satisfaction of the elderly in the process of robots taking care of the elderly,a service object satisfaction model is constructed.An agent based on the satisfaction model of service object is designed to realize the scheduling of the coordination framework,the agent makes personalized planning for the tasks of the elderly and reasonably distributes the system resources by considering the "personal factors" of the elderly and the real-time status,location and other information of the robots.Make the robots help the elderly to complete the task while maximizing the overall satisfaction of the elderly.Thirdly,the simulation and analysis of the decision made by the agent under different circumstances are carried out.The results show that the agent's decision is reasonable.Finally,an optimal solution for large-scale task conflict is proposed.Particle swarm optimization algorithm combined with clone mutation operator is used to solve the task execution scheme quickly,which improves the response speed of the system.