Research On Task Allocation Of Multi Robot System For The Intelligent Manufacturing

The trend of manufacturing industry is more and more intelligent around the world in the future,which has attracted wide attention of the world, such as the United States advanced manufacturing industry national strategy, Germany "industrial 4.0", "Made in China 2025" etc.. Robot is the important cause of the future intelligent manufacturing equipment, it is a kind of important and special carrier to realize intelligent manufacturing. As a result, multi robot system has become one of the hot topics in the field of manufacturing, and task allocation is a key problem in the research field of multi robot system. But most of the existing multi robot task allocation methods are based on the specific application scenarios. It is suitable for a specific environment, the general purpose is poor, the number of robots is relatively small, and most of them are homogeneity robots. And intelligent manufacturing mode, the production environment is dynamic and uncertain, and the number of robots is huge, and it is a heterogeneous robot. In the face of actual manufacturing environment, there are also flexible and real-time requirements.In view of this, this paper presents a formal description method of the task and robot, based on the analysis of the multi robot system, and proposes a general, standard and suitable task decomposition method based on robot capability and task characteristics. Then, the task of the large scale heterogeneous multi robot system is optimized and the corresponding management suggestions are given, in order to make the multi robot system work in a reasonable and efficient way.The main research contents and innovation points are as follows:(1) Firstly, this paper introduces the research background of multi robot system based on intelligent manufacturing, and introduces the research status of multi robot task allocation problem, and introduces the concept of role assignment. The paper presents the necessity and advantage of the concept of role assignment in multi robot system task allocation. E-CARGO model is first introduced into the research field of robot task allocation. This paper presents the research framework of multi robot system task allocation based on role.(2) We have studied the task decomposition based on the role and capacity classification. Task decomposition is the early preparation of task allocation in complex systems. But most of the existing task decomposition is to divide the task into two stages, the existence of human being separated from the defects. In this paper, a general method is proposed, which is based on the combination of the robot’s capability vector and task role requirements. The method takes into account two aspects, such as task role requirements and robot’s ability. In this method, the task is described by the formal description of the task, and the capability of the robot is constructed from the point of view of capability classification. Finally, the task decomposition method based on the role and capacity constraints is proposed.(3) In this paper, we design a centralized task allocation based on role. Based on the task decomposition, the task assignment of multi robot system is studied. If the condition is to grasp the task and the global information of the robot, the centralized optimization allocation, which mainly includes three parts, such as fixed role assignment, flexible role assignment, dynamic role adjustment, etc..(4) Finally, we study the distributed task allocation based on role. Intelligent manufacturing mode of market competition, enterprises are facing complex and varied production tasks. This part of the main content is assumed to not have the global information, considering the learning ability of the robot, and fully exert its social, intelligence, to carry out the role based multi robot system distributed task allocation. Five sets of experiments were conducted to investigate the effects of task allocation, the number of robots, the number of robots, and the change of space. We discuss the factors that affect the role assignment of multi robot system, and finally get the optimal allocation scheme and the corresponding management implications, and provide methods and suggestions for the effective organization and management of multi robot system.