Target tracking of the robot fish based on adaptive fading Kalman filtering

To solve the problem that underwater robot-fish can not track the dynamic targets accurately in water, an underwater target tracking method based on adaptive fading Kalman filter (AFKF) algorithm is proposed in this paper. The output of AFKF is used to locate the target under water in real-time, which helps the robot-fish calculate the optimal tracking path. Applying the AFKF, the parameters of the target are adjusted continuously, thus the filtering divergence is prevented. The simulations show the correctness and effectiveness of the algorithm. We then apply the algorithm to the 5 VS 5 competitions of China Underwater Robot-fish games, the experimental results on the two-dimension simulation platform validate that the adaptive fading Kalman filtering significantly improves the efficiency and accuracy of the tracking process.