Flexible tail motion estimation by electromagnetic induction for a biomimetic underwater vehicle

In order to mimic a real fish, the back bracket of a biomimetic-underwater vehicle's (BAUV's) tail fin is composed of aluminum spacers with a flexible aluminum plate as the tail spline. The sway of the tail fin is driven by a motor and its motion is impeded by the water such that the rotation angle of the tail fin could not be calculated simply by a coordinate conversion. For precise tail fin control in a complex environment, the motion information of the tail fin is required. In this work, the development and setup of a magnetic field system is presented. The system is composed of a magnetic transmitter fixed on the robot fish's body, and three magnetic receivers located along the fish tail, as a moving coil structure. Receiver and transmitter circuits were designed based on the resonance theory. When relative displacements and relative angles appear between the transmitter and receiver, the receiving side displays specific voltage signals, which are used to quantify the position and orientation of the moving coils. The shape of the BAUV tail fin is approximated by a body-spline equation. Three receiving coils information were used to interpolate the body-spline of the BAUV tail motion. A database that relates sensor data with the sensed positions of the BAUV tail fin is established. With the database, the system used the K-nearest algorithm to classify the digital signals that are conversed by the FPGA A/D function into the sensors positions, such that the motion of the tail fin can be estimated.