Force control of series elastic actuators using sliding mode control theory

Force control in robotic actuators is crucial for stability, safety, and adaptability in contact environments. Series elastic actuators (SEAs) offer advantages such as improved force control, shock absorption, and reduced noise but introduce challenges related to bandwidth and precision. This paper presents a force control strategy based on sliding-mode control (SMC) to enhance robustness and stability in dynamic environments. By treating environmental dynamics as disturbances, the approach ensures stable interaction. A chattering-free implementation is proposed, leveraging simplified task models to maintain high performance. Theoretical stability is established within the sliding mode framework, with safety ensured through the careful design of the sliding variable. Analysis demonstrates the method’s ability to generalize across various interaction scenarios, including stiff contacts, purely inertial loads, and soft materials. The adaptability arises from model parameters with interchangeable physical meanings, allowing seamless application across different environments. However, as environmental stiffness decreases, precision in convergence may degrade. To address potential precision loss in low-stiffness environments, a sliding-mode robustification technique is introduced. Simulations validate the proposed controller’s performance, demonstrating convergence consistency with theoretical expectations. The findings highlight the method’s effectiveness in achieving stable, adaptive force control across diverse conditions.