Direct Force Compensation-Based Adaptive Hybrid Impedance Control for Dynamic Interaction Environments
The primary control objective in robotic interaction tasks has shifted from trajectory tracking accuracy to interaction force regulation. Impedance control enables compliant interaction by shaping the robot’s force response through a target impedance model. However, the dynamic response of conventional impedance control is fixed during parameter design, limiting its adaptability to varying interaction environments. To address this issue, a direct force compensation–based adaptive hybrid impedance control (DAHIC) method is proposed. By introducing an adaptive force compensation factor based on the force tracking error, the proposed controller achieves faster response speed while suppressing overshoot and reducing steady-state error, especially in tracking higher-order signals. The allowable range of controller parameters is derived via closed-loop stability analysis, and an adaptive update law is designed accordingly. Furthermore, the proposed adaptive force compensation is integrated with hybrid impedance control to form a six-dimensional task-space compliance controller. The numerical simulations conducted in the single-force subspace, together with the comparative experiments performed in the six-dimensional task space involving constant- and variable-curvature contact surfaces as well as abrupt variations in contact stiffness, collectively validate the improvement in force tracking performance achieved by the proposed algorithm. Note to Practitioners—In robotic interaction tasks, such as assembly, surface contact, and human–robot collaboration, achieving stable and accurate force regulation in uncertain environments remains challenging for conventional impedance controllers with fixed parameters. This paper proposes a direct force compensation–based adaptive hybrid impedance control method to improve force tracking performance under varying interaction conditions. By adaptively adjusting the force compensation according to tracking error, the controller achieves faster response, reduced overshoot, and improved steady-state accuracy, and can be naturally extended to six-dimensional task-space compliance control. The stability of the closed-loop system is ensured through analytical parameter constraints, providing practical guidance for controller design. Practitioners may find this approach useful for enhancing compliant interaction performance in real-world robotic applications, while practical implementation requires appropriate force sensing, real-time computation, and careful parameter tuning.