Multi-objective evolutionary design of central pattern generator network for biomimetic robotic fish

• DOI: 10.1007/s40747-022-00883-7
• Publication Date: 2022-10-07T02:05:21Z
• Authors: Wei Kun Li; Hao Chen; Wei Cheng Cui; Chang Hui Song; Lin Ke Chen
• Keywords: Biomimetics; Digital pattern generator

Abstract Fish-inspired motion is an important research area with many applications in real-world tasks such as underwater vehicles or robotic fish control design. Owing to robust, smooth, and coordinated oscillatory signals generated by Central Pattern Generators (CPGs) for locomotion of robots multiple degrees freedom, CPGs are the most versatile solution systems, especially fish. However, tuning central pattern generator parameters difficult complex mechanical system designs. Besides, current CPG-based methods only consider one aspect (e.g., speed), which widens gap between theory practice Also, it may affect practical applicability designed model a certain extent. This paper addresses this problem constructing multi-objective evolutionary design network proposed biomimetic A new CPG help swim efficiently. In addition, efficient algorithm our previous work also applied assist obtaining faster-swimming speed, good stability head, higher propulsive efficiency simultaneously. Considering that result optimization set non-dominated solutions rather than solution, screening method based on fuzzy adopted decision-makers selecting appropriate solution. Based this, constructed. The simulated compared seven well-known algorithms series After that, validated extensive experiments actual Simulations demonstrate model’s effectiveness performance, when has been real