Background Microspine grippers are critical for wall-climbing robots and drones to attach to rough surfaces in complex environments. Existing designs focus on structural practicality but overlook rational dimensional optimization and accurate modeling of adhesion force. One-degree-of-freedom (1-DOF) grippers also exhibit poor collision resistance and low grasping stability on variable-roughness surfaces. Methods To address these shortcomings, a two-degree-of-freedom (2-DOF) biomimetic microspine gripper inspired by the retractable claw structure of feline paws is proposed. A microspine stiffness model based on Castigliano’s Second Theorem, combined with a rough surface model and a gripper statics model, is established to quantify adhesion force. The performance atlas method combined with dimensionless parameter processing is adopted for dimensional optimization, where the top 25% of adhesion force data in each subplot is defined as the high-performance region to identify the optimal parameter combination. Results Experimental validation shows the stiffness model has a corrected relative error of only 4.5%. The optimized gripper achieves stable adhesion of 30 N and 22.7 N on rough asphalt and smooth stone surfaces, respectively, with a 95% grasping success rate on variable-roughness surfaces. Compared with our self-developed 1-DOF gripper, the proposed design effectively reduces collision-induced microspine damage and significantly improves grasping stability and environmental adaptability. Conclusion This work provides a theoretical and optimization framework for microspine gripper design, and the biomimetic design strategy offers new insights for the development of high-performance robotic attachment components.
Wen et al. (Thu,) studied this question.