Experimental and theoretical verification of TLBO and PSO for solving the inverse kinematic model of continuum robots

Abstract

This paper presents a comprehensive exploration of two meta-heuristic optimization techniques, Teaching Learning-Based Optimization (TLBO) and Particle Swarm Optimization (PSO), applied to solve the inverse kinematic problem of continuum robots. The study encompasses both theoretical investigations and realistic simulations, including tracking a spiral trajectory and utilizing real measurements to follow a trajectory. TLBO demonstrates exceptional precision in solving the inverse kinematic problem for continuum robots, consistently outperforming PSO in terms of accuracy. On the other hand, PSO showcases notable advantages in terms of computational efficiency, exhibiting faster convergence and reduced time consumption. The research findings suggest promising avenues for the application of meta-heuristic approaches in real-world scenarios involving continuum robots, particularly in domains such as medical devices and industrial automation. However, the challenge remains to develop modified algorithms that strike a balance between accuracy and efficiency to address the diverse requirements of practical applications in this field. Nevertheless, the versatility of meta-heuristic methods in handling complex robotic systems offers exciting prospects for the future of continuum robotics.