Dynamics Analysis And Control Of Spiral Motion Of Robotic Fish

As a new type of underwater vehicle,underwater bionic robot has many advantages,such as high propulsion efficiency,low noise and little disturbance to the environment,so it has been widely studied.Because of the highly nonlinear dynamic model of robotic fish and the uncertainty of fluid dynamics,the research of motion control has always been the core technical problem in this field.This paper takes the bionic robotic fish propelled by two fins in the laboratory as the object,and studies the method to improve the motion control accuracy of robotic fish and enable the robotic fish to cope with complex environment under the mode of pectoral fin/ caudal fin coordinated propulsion.The specific contents are as follows:(1)Based on the experimental data,a dynamic model of the two-degree-of-freedom pectoral fin/flexible body-caudal fin propulsion robotic fish was established.The pectoral fin is the auxiliary airfoil,which can improve the stability and cornering ability of the robot fish.Pectoral fins have two degrees of freedom: flapping and swinging.The combined movement of the pectoral fins not only reduce the resistance of the pectoral fins in the process of movement,but also improve the speed of diving,rising or turning of the robotic fish.Therefore,the force and moment generated by the pectoral fins are calculated based on the Angle of attack theory.The flexible body part is made of the three joints,which provides most of the propulsion force when the robotic fish swims.In order to obtain more accurate expressions of force and moment,the relationship between swinging frequency and force is obtained by fitting several groups of experimental data.Finally,the force and moment generated by the motion of the thoracic/caudal fin are converted into the dynamic equation in the spatial coordinate system by the rotation matrix.Three different spiral diving methods are proposed,and the angular velocity curve is obtained based on the established dynamic model.The experimental and simulation results show that the data aided model of robotic fish has higher accuracy than the traditional model.(2)Using the established CPG network,a discrete sliding mode controller was designed to realize the angular velocity tracking of the bionic robotic fish.The sliding mode controller adjusts the oscillating frequency of the flexible body part and pectoral fin online according to the real-time data transmitted by the sensor,so that the robotic fish can reach the target angular velocity in the pectoral and caudal fin cooperative mode.Two sets of simulation and experiment schemes are designed,and the results show that the steering angular velocity of the bionic robotic fish using the discrete sliding mode controller can converge to the expected value.Among them,the error value of low-speed experiment is ±0.03 rad/s,and the error value of high-speed experiment is ±0.04 rad/s.