Learning From Fish: A Two-Stage Transfer Learning Method for a Bionic Robotic Fish

• Published in: IEEE transactions on automation science and engineering Vol. 22; pp. 18796 - 18808
• Main Authors: Yu, Fuyang, Wu, Zhengxing, Wang, Jian, Yu, Lianyi, Feng, Yukai, Tan, Min, Yu, Junzhi
• Format: Journal Article
• Language: English
• Published: IEEE 2025
• Subjects: bionic robotics; dynamic trajectory tracking; Dynamics; Feature extraction; Fish; Motion capture; motion learning; Oscillators; Robot kinematics; Robotic fish; Sports; Trajectory; Trajectory tracking; Transfer learning
• ISSN: 1545-5955; 1558-3783
• DOI: 10.1109/TASE.2025.3591314

Directly learning the swimming behaviors of real fish can significantly enhance the swimming performance of bionic robotic fish. This paper presents a novel transfer learning method based on a dynamic trajectory control approach for the robotic fish to learn swimming skills from real fish. First, we develop a fish motion capture system and a crucial motion extraction approach to realize precise decomposition of fish motions and collect abundant meaningful features from a snakehead fish as pre-training data. Next, we construct a two-stage transfer learning method based on Deep Deterministic Policy Gradient (DDPG), including an offline and an online stage. Specifically, in the offline stage, the obtained pre-training data is processed for experience learning within a DDPG-based network, whereas in the online stage, a dynamic trajectory tracking method is utilized to refine the robotic fish's motions in real time based on the learned strategies. Experimental results on a self-developed four-joint robotic fish show that the proposed method effectively extracts and transfers biological motion features into the motion control of the robotic fish. Compared to the conventional CPG method, the proposed approach exhibits stronger acceleration capabilities and more efficient swimming, resulting in enhanced maneuverability of the robotic fish. Overall, this approach provides a technical foundation for bionic robotics to learn from nature. Note to Practitioners-The motivation of this work stems from the need for more adaptable control of robotic fish swimming behaviors to improve their maneuverability. Traditional dynamic trajectory tracking methods lack the flexibility to adjust motion characteristics according to specific requirements. In contrast, the swimming motions of real fish offer valuable insights for controlling robotic fish. This paper proposes a two-stage transfer learning framework that extracts action features from biological fish movements across various conditions, and then transfers these features to robotic fish for optimization in real-world environments. The knowledge gained from biological motion characteristics provides effective guidance for controlling robotic fish swimming behaviors. In practical scenarios, the proposed method enables control of robotic fish movements based on factors such as acceleration requirements and oscillation constraints, offering enhanced flexibility in acceleration and cruising performance.