Simulation-In-The-Loop Optimization of Obstacle-Aided Locomotion in Snake-Like Robots

Abstract

This study investigates the optimization of obstacle-aided locomotion in snake-like robots through a simulation-in-the-loop framework.A dynamic model of the robot is developed in CoppeliaSim to evaluate contact forces interactions during navigation in environments with varying obstacle densities.These interactions are used to optimize the robot's motion trajectory with minimal actuator energy while penalizing contact forces.The robot's motion is parametrized by a set of serpentine gait parameters, optimized using Particle Swarm Optimization to exploit beneficial obstacle contacts.Results reveal a trade-off between obstacle utilization and energy efficiency, demonstrating the potential of obstacle-aware trajectory planning for enhanced locomotion performance.