| Because of the importance of ocean in military and resource, the underwater robots become a hot topic of international research in recent years.The propulsion method of fish has the advantage of high speed, high efficiency, excellent maneuvering performance, low noise etc. Therefore, the biomimetic robotic fish becomes the focus of underwater robot research.However, now the research of biomimetic robotic fish focuses on the movement mechanism and the mechanical structure, there is little research on the path planning of robot fish. This paper applies genetic algorithm and dynamic programming to path planning of robotic fish successfully and achieves good effects.In this paper, the main research work and the current research situation at home and abroad are introduced first. Then an introduction of the multi-robot control system is made, containing the hardware environment and software environment. After that, the principle of genetic algorithm is introduced.The steps of path planning based on genetic algorithm are as follows. First, the environment is modeled by grid methods; then an optimal robot path in the grid environment can be obtained via the genetic operations, such as reproduction, crossover and mutation etc. Velocity control algorithm and direction control algorithm can be designed in terms of fuzzy control and subsection control respectively, so as to make actual path and the path obtained via genetic algorithm as near as possible. Finally the computer gives corresponding control instructions to the robotic fish. Experimental results show the effectiveness of the proposed method.Since the characteristic that the robotic fish can't track linear motion, the robot path can be seen as several circular arcs, thus robotic fish dynamic programming algorithm is designed. Then not only an optimal circular arc path, but also the speed and the direction of each state can be obtained. The experiment on the robotic fish shows the effectiveness of the proposed method. |
PDF Full Text Request