Central Pattern Generator Control for a Biomimetic Robot Fish in Maneuvering

摘要

Fish has excellent maneuvering abilities in water, such as fast turning speed, small turning radius and large acceleration, which motivates researchers around the world to study its swimming mechanisms and create some biomimetic autonomous underwater vehicles (AUVs) for practical applications. In this paper, we observe the motion of one black carp in the length of 80mm and focus on one of its turning modes. Then, the turning pattern of this mode is characterized by four parameters, i.e., the amplitude, the frequency, the offset and the time ratio between two phases forming one cycle, i.e., the beat phase and the restore phase. Based on the fact fish swimming is produced by Central Pattern Generator (CPG), we propose a modified CPG model which can generate rhythmic signal matching the carp's turning pattern. Moreover, this CPG model is also capable for cruising and the transitions among different swimming modes, which really makes controlling of a robot fish convenient and improves its performance. Finally, experiments are conducted to examine the effectiveness of this control method.