Fuzzy Control Of Biomimetic Robotic Fish Based On Genetic Algorithm

Bionic robot fish is a combination of mechanical, electronic and other multi-disciplinary disciplines to design the water robot. It simulates the fish in the ocean of real life and forms the propelling mechanism of robot fish swimming. It has very important application value in salvage rescue and seabed detection. Path planning research is the basis of coordinated control of bionic robot fish. The research object of this paper is bionic fish and the bionic robot fish control system is described in detail. The global vision subsystem is researched and analyzed carefully,which lays a foundation for path planning of bionic robot fish. Then, the environment modeling of the bionic robot fish was carried out, and the path planning and its method optimization were completed by using the genetic algorithm. The fuzzy controller was designed and experimentally verified on the bionic robot fish control system.In this paper, the research value and significance of robotic fish are discussed firstly.Also, the research and development of robotic fish in China and abroad are summarized.The main work of the paper is summarized. Secondly,the paper discusses the establishment of robotic fish control system,which paves the way for the path planning research in the following context. Thirdly, the basic characteristics, problems and methods of the planning path based on the path planning for robot fish are introduced in this thesis. The method of path planning in this paper is determined,and a hybrid algorithm based on grid method and genetic algorithm is proposed. Fourthly, the content and idea of genetic algorithm are introduced. The obstacle avoidance algorithm (OAA)and the differentiation algorithm (DA) are proposed to generate the initial population and check the path so that all individuals appearing in the first generation are feasible. In the improved genetic algorithm, some special genetic operators are introduced, such as refinement and deletion operators based on domain heuristic knowledge. These operators are used to plan and form the best path of bionic robot fish. Finally, the fuzzy controller is designed for the fish in order to achieve the path planning. The input and output as well as the fuzzy rules are established. Besides, both fuzzification and defuzzification are studied. In order to verify the proposed algorithm, the experiment of the smooth control of the robot fish on the system platform is carried out.