Crustal movement modeling and analysis are the major task in the earth system science. Crustal movement simulation system is a new kind of experimental instrument used for physically simulating geological activities in the laboratory environment. Obviously, it has many advantages to use such a system: greatly shortened time and space scales of the process of crustal movement, completely safe working environment for researchers and more convenient method of observations. Although the simulation technology is a field in which a significant amount of research has been conducted, little work has been done in the combined area of crust science and automation technology. The hardware system consists of mechanical system, control systems, communications systems and central monitoring and management system. In the hardware point of view, it is a distributed structure using sixteen or more square modules and a central computer. Each square module represents a crustal plate. By driving these modules, we can imitate the Earth's crustal movement in a laboratory environment. Several control issues arise due to the mechanical structure of the system and due to the peculiar nature of the tasks. The paper is supported by the National High Technology Research and Development Program of China (863 Program) and the Major Scientific Equipment Development Project of Chinese Academy of Sciences. This thesis is focused on the crustal movement simulation system and selects the study on multi-robot technology based control and coordination algorithm as its main topic. In this paper, the main work and contributions are: 1. An embedded open-controller is designed based on virtual machine technology. In this way, the control logic can be completely isolated from the underlying platform. So it achieves many outstanding features such as openness, scalability, portability and platform independence. In addition, the controller can support multiple programming languages. 2. An efficient path planning algorithm based on immune evolution is proposed. This algorithm makes the simulation system in the best fit the requirements of a given terrain only using a finite number of flat modules in three-dimensional space. And on this basis, particle swarm optimization is introduced to further improve the convergence speed. Adaptive fuzzy control algorithm is selected for the motion control algorithm. 3. A decentralized multi-robot flocking algorithm is proposed and implemented, which enables c...
修改评论