Enhancing snake robot stiffness in confined spaces: Pose optimization and motion strategy for pose adjustment

Snake robots excel in shape adaptability within unstructured and confined environments. However, their slender structure inherently limits end-effector stiffness, restricting them to ultra-light-load operations tasks such as inspection, welding, or minimally invasive surgery. To address this, we propose a bio-inspired, three-stage motion strategy integrated with pose optimization: 1) obstacle-avoiding navigation using serial-mechanism motion modes and follow-the-leader control; 2) precise end-effector positioning via parallel-mechanism motion modes; and 3) micro-scale “shape-shoulder” pose adjustment to maximize task-space stiffness. By decoupling stiffness optimization from initial navigation, our method reconciles high stiffness with safe obstacle avoidance. We formulate hybrid kinematics and stiffness-based pose optimization models, implemented on a modular, underactuated variable-geometry truss-type snake arm. The optimization framework incorporates a penalty-based obstacle avoidance strategy directly into the objective function, ensuring smooth gradient convergence and enabling exploration of stiffness-optimal configurations near obstacles while maintaining required safety margins. Case studies in an aircraft wing box and an enclosed nuclear reactor demonstrate the robot’s significantly enhanced stiffness in confined spaces, expanding its potential for high-stiffness manipulation in constrained industrial environments.