Research On Collaborative Task Planning Method Of Robots

The application of robots reduces people's labor intensity and the danger in dangerous and unknown environment.Especially,the application in war has promoted people's research on robot technology.Multiple robots have obvious advantages over single robots when they perform spatially dispersed and time continuous tasks.With the development of computer science and artificial intelligence technology,people have more and more expectations for robots to perform complex tasks.The cooperative mission planning of multi robot systems has attracted wide attention and research.Task planning mainly includes two aspects: task allocation and path planning.At present,the research of multi robot path planning is more thorough.But the complex mission planning and organization of multi robots belongs to the top-level design.It involves operations research,game theory,artificial intelligence and other disciplines.The research in this field is difficult,most of which do not have very strong application background.Starting from the difficulties of the research,the paper takes the construction of complex problem models with applied meaning as the main line.On the basis of studying the basic theory of robot planning.It aims at the US Air Force mission planning system framework.The model of cooperative attack is studied as the carrier.The overall goal task allocation and sub goal cooperative search problem are studied.Firstly,the basic theory of robot task planning is studied.The research focuses on two aspects: system structure and location technology.Architecture is the first consideration in the top-level design of multi robot system.It is of great inspiration to the successful establishment of task planning problem modeling.Robot localization technology is the basic and underlying problem of robot system.Research on it can lay a foundation for future application of problem models.The choice of location technology has a direct impact on the overall design of specific mission planning systems.Secondly,a robot cooperative task planning model is proposed.The model of cooperative attack is modeled.In the four aspects of mission objectives,available resources,environmental threats and task demands,specific problems are modeled.The common planning method of centralized allocation method is studied.The classical optimal allocation method and integer linear programming method are combined with model problems respectively,and two new specific models are proposed.The simulation is carried out with two planning methods.The simulation results show that the proposed problem model is effective and feasible in solving complex cooperative attack problems.Finally,for a sub task in task assignment,that is,the target search task.A distributed swarm robot cooperative search method is used to solve the task.Those robots should be divided into some sub-swarm by job allocation,so that each sub-swarm can cooperatively work focusing on its desired target.Each sub-swarm searches for its target by evolutionary algorithm.In order to improve the cooperative search ability of sub-swarm,a new particle swarm algorithm based on auxiliary orientation improvement is proposed to enhance the efficiency and accuracy of the search.In the simulation,the general roaming algorithm consumes a longer time to find the target,and reduces the overall efficiency of the system.In order to solve this problem,a spiral roaming method based on artificial potential field is introduced to improve the global search capability.The simulation results show that the search efficiency is improved significantly by using this method.