Research On Intelligent Control Of Multi-joint Bionic Robotic Fish Swarm

Bionic robotic fish has become a focus of research for scholars due to their potential applications in environmental surveying,ocean resource exploration,search and rescue,and other fields.Among them,the intelligent control of multi-jointed bionic robotic fish swarms is an emerging research field that has received increasing attention from scholars.With the deepening of research,multi-robot fish can complete tasks that cannot be accomplished by a single robot fish,such as multi-fish handling,multi-fish search and rescue,multi-fish reconnaissance,multifish mine clearance and other tasks,so multi-robot fish is taken as the research object of this project.Therefore,the subject of this study is multi-bionic robotic fish.The intelligent control system of multi-bionic robotic fish requires the integration of various intelligent algorithms,including swarm algorithms,path planning algorithms,path tracking algorithms,and so on,in order to achieve the goal of multiple bionic robotic fish gathering together to complete certain tasks.Firstly,establish the motion model of the robot fish and propose its motion control strategy.The first step is to analyze the kinematic model of fish movement to understand the power source of the robot fish.The second step is to derive the dynamic and kinematic equations of the robot fish through the robot fish linkage model and Euler-Lagrange equation.The third step is to propose a control strategy,which is to change the input torque of the three motors at the joints of the robot fish to control its speed and direction.Finally,MATLAB simulations are also performed for the designed machine fish model.Secondly,the distributed control approach is used to design a path planning algorithm for a school of robotic fish.In order to make the fish school move towards the target point without collision,avoid obstacles,and eventually reach the target area,a combination of swarm strategy and decentralized navigation function is proposed for the path planning algorithm.This approach ensures that the fish school moves towards the target area while keeping a safe distance from each other and obstacles.MATLAB software is used to simulate the designed algorithm,and the reliability and universality of the algorithm are verified by changing the complexity of the environment.The superiority of the algorithm is also verified by comparing it with a centralized algorithm.Thirdly,in order to enable the robotic fish to converge to the desired path even under uncertain disturbances in water,the active disturbance rejection control algorithm is used to design the path tracking of the robotic fish,which has a stronger anti-interference ability.MATLAB simulation software is used to compare the active disturbance rejection control algorithm with the PID and fuzzy PID algorithm,and the results verify the advantages of the active disturbance rejection control algorithm,such as fast convergence speed and strong antiinterference ability.Finally,experiments are conducted in the underwater robotics laboratory.Two types of experiments are conducted to validate the algorithm designed above using a multi-jointed robot fish in the laboratory as the experimental object.The first type of experiment requires the three robotic fish to gather and swim towards a fixed small ball until they reach near the ball.The second type of experiment requires the robotic fish to swim on the desired path despite the disturbances on the water surface.Two groups of experiments are conducted to validate the feasibility of the designed cluster path planning algorithm and path tracking algorithm.The advantages of the algorithm are summarized and the outlook for future work is proposed.