Study On Motion Control Of Carangiform Robotic Fish Synergistically Propelled By Pectoral Fins And Caudal Fin

In order to make Carangiform robotic fish have different swimming abilities to grapple with complex tasks,the research on the motion control has always been a research hotspot in the field of Carangiform robotic fish.However,the existing researches on the motion control of Carangiform robotic fish are mostly concentrated under the propulsion mode which uses body and caudal fin,while the researches on the motion control under the propulsion mode which use pectoral fin cooperated with caudal fin are less.Based on the Carangiform robotic fish platform independently developed by the laboratory,this dissertation has done the following research on the motion control of Carangiform robotic fish:(1)For the robustness problem in motion control of biomimetic robotic fish,the Hopf oscillator model of the central pattern generator(CPG)is used to establish the CPG controller of each joint of the robotic fish.The CPG control network is constructed through the coupling matrix and the rotation matrix.The parameters of CPG to achieve basic swimming modes of robotic fish are determined by the combination of simulations and experiments.The results of simulations and experiments show that the swimming modes of the robotic fish possess good robustness.(2)In order to consider dynamic performance and steady-state performance in the process of depth control at same time.Combining the established CPG network,the depth control of the robot fish is divided into the approaching stage and the cruising stage and the fuzzy controller is designed respectively to realize the depth control of the robotic fish.In the approaching stage,the cooperative motion of the pectoral fins and caudal fin is adapted to quickly approach the desired depth,and the approaching fuzzy controller is used to adjust the bias and amplitudes of the pectoral fins in real time.In the cruising stage,the cruising fuzzy controller is adapted to adjust the attack angle of pectoral fin to keep the robotic fish at the desired depth and reduces steady-state errors.The effectiveness of the designed controllers are verified by simulations and experiments.It can be seen that the performances of the designed controllers are improved compared with other depth control method of robotic fish.(3)The process to tracking stationary target which from detecting the target to approaching the target is completed by decomposing the tracking problem of the robotic fish into the detecting stage and the approaching stage.In the detecting stage,the hierarchical detecting method which is based on the completed depth control is used to search the target.Compared with other detecting methods,this method is not efficient,but it simplifies the design process of the controller.In the approaching stage after the target is found,the pitching motion and yawing motion are simultaneously controlled by two parallel fuzzy controllers to make the robotic fish to approach the target.The simulation results show the effectiveness of the tracking method.