| With the development of robotics and with the growth of social demands, we are more interested in multi-robot system than in single robot system. On the one hand, because of the complexity, when it is difficult for single robot to complete tasks we want to complete them by the coordination and cooperation of multi robots. On the other hand, we need to improve the efficiency of system by the coordination and cooperation of multi robots. Compared with a single-robot system, a multi-robot system involves some advantages as follows: wider application fields, inherent parallelism, higher efficiency, robustness, easier and cheaper to build, distributed sensor and cooperation, scalability, beneficial to study swarm intelligence. Accordingly, more and more researchers are interested in multi-robot systems. In the past two decades, multi-robot systems have improved greatly in theory and application. Nowadays, MRS is applied in various fields, such as the cleanup of toxic waste, planetary exploration, search and rescue missions, surveillance, cooperative transportation. With the increasing requirements, MRS will be applied to broader fields. In following years, the major research directions on MRS will focus on enlarging the number of robots, improving the cooperation ability, studying the swarm intelligence. Environments of MRS will be changed from certain, structured and static to uncertain, unstructured and dynamic. The results of study will become real applications. Therefore, it is very important to research MRS used into the complex and dynamic environments. This paper studies multi-robot motion coordination in a complex and dynamic environment based on forage task. The main research work about this dissertation is as follows: 1. Build a distributed multi-robot simulation system based on C/S structure. Because robot usually is expensive, simulation is necessary to research the robot system. After the algorithm is verified in simulation system, it can be applied to a real system to protect robots. Especially, simulation is necessary for a multi-robot system in complex and dynamic environments. Accordingly, the development of a distributed multi-robot simulation is discussed in this paper. Server is a central station and responsible for a user interface and setting parameters of system. In addition, central station serves as a global monitor to show the executive state of the system. The client uses multiple threads to instantiate executive robots and each thread denotes a robot. The simulation system is developed based on Agent techniques and programmed with JAVA so that it can run on various OS and can validate different algorithms. 2. Extend KQML according to KQML syntax. Some performatives and corresponding semantic parser are defined in JAVA. The extended KQML is applied to the multi-robot simulation system and complete communication between robots. Due to utilizing KQML protocol, the openness and compatibility of the simulation system is improved greatly. 3. Considering the scale of task and the optimality of the allocation, we bring forward or investigate OD algorithm, Hungary Algorithm and gene algorithm in this paper based on two-dimensional task allocation problems model. Task allocation is to find a most reasonable allocation project in many methods. It behaves to find a result satisfied the given restricted conditions and to make the result of object function maximal or minimal. The OD algorithm in this paper is to make every robot get the most suitable task in view of the optimization. However, this algorithm is only suitable for small-scale task allocation problems because of its large scale of operation. The Hungary...
|
PDF Full Text Request