Multi-Robot Path Programming

Because of the diversity and complexity in social requirements, higher performance is expected in robot technical development and conventional single-robot work mode has been not fulfilling the reality demand. In order to improve work efficiency and action ability, it's necessary to research multi-robot technology. In comparison with single-robot, multi-robot exhibits more intricate features, such as control scheme, environment perception, dynamic decision, cooperation pattern, information communication and so on. Path programming is a hard work for a multi-robot and it must be a key research aspect in multi-robot development. The purpose of this paper is mainly to explore the intelligent path programming of multi-robot in static and dynamic unknown environments. The major work of this paper including:1. Summarizing the current research achievement and trend of multi-robot, making a brief introduction on path programming of robot, research situation in national and international and conventional methods in global and local path programming of robot. Finally, a perspective on multi-robot path programming development trend is made. 2. Multi-robot path programming in different work environment is studied, including the path programming in global static known environment and local dynamic unknown environment. For the multi-robot path programming in known environment with barrier, a method philosophy based on the rolling optimization strategy and genetic algorithm for dynamic path programming is proposed. It can produce a series of moved space window by rolling strategy, conduct on-line programming in the window, measure the real time window environment information through the own sensors of robot, and solve out the optimal trajectory in every shifting window by genetic algorithm. The simulation results indicate that the method proposed can find out an optimal trajectory for objective detection, which overcome the disadvantage of higher complexity in conventional local path programming methods. Furthermore, the method improves the operability of algorithm, accelerates convergence rate, and fulfills the real time demand for robot path programming under unknown complicate dynamic environment.3. For the multi-robot path programming in global known environment, a particle swarm optimization algorithm is devised. In order to make multi-robot coordinate perfectly in work environment, it's necessary for them to communicate with each other, and adjust their behaviors. When path programming using particle swarm algorithm, the robots can communicate with each other, and then compare location and velocity of current and future, obtain a global optimal trajectory finally. The simulation results demonstrate that the method proposed can make coordination between robots perfectly, improve work efficiency, and meet the experiment specification.4. Path programming is very important for multi-robot. It will be face some challenges in solve path programming, such as cooperation between robots, object detection, increase in number of robot and enhancement in space complexity. Object detection method is investigated in this paper. At the same time, the merit and demerit of different control structures are analyzed. The results shown that control mode is more stable and information communication more fast for multi-robot.Finally, the contribution of this paper is summarized and deficiency in this study is analyzed, and then remain work of path programming for multi-robot is pointed out and perspective of path programming for multi-robot is made.