MODELING AND MOVING OBJECT TRACKING CONTROL OF A THREE LINK ROBOTIC FISH

Robotic fish have attracted the attention of many researchers over the last decade due to their diverse applications. The dynamic modeling and control of fish robots is an important and challenging problem. The modeling of robo-fish is useful in the design, manufacturing, efficiency estimation and control of fish robots. Presently, no exact analytical models are available that can predict the propulsive forces of a three-link robotic fish. In the present study, analytical dynamic modeling of a three-link robotic fish and its control to track moving objects is presented. Perhaps the best known theory for swimming is Lighthills Elongated Body Theory (EBT), which has been used to study anguilliform and carangiform propulsion. In this paper, Lighthills large amplitude Elongated Body Theory has been used to obtain the thrust forces for a three link robotic fish. Due to the complexity of the resulting dynamic model,utilization of nonlinear and robust controllers becomes impractical. In order to track moving objects, a fuzzy controller and a Brain Emotional Learning Based Intelligent Controller (BELBIC) have been proposed. Two different types of controller are designed using BELBIC and fuzzy control methods. Two controllers, one for the direction of motion and the other for tail swinging frequency, are designed in each case. The directional controller directs the fish robot towards the target, while the frequency controller adjusts the distance between the target and the robot. Finally, the dynamic equations of the robotic fish have been simulated and the performance of the proposed controllers is investigated. The simulation results show that the dynamic equations are able to simulate unsteady effects, and the robot is able to track a moving object properly using the proposed controllers. The results also reveal that the designed controllers provide good robustness to parametric uncertainties. Finally, the BELBIC controller is shown to have better performance compared to the fuzzy controller.