An adaptive finite-time control approach for telerobotic systems with time-varying output constraints

Aiming at the high-precision synchronous control challenge of teleoperated robotic systems with communication delays and position constraints, this study proposes a groundbreaking finite-time position tracking control framework. It achieves direct constraints on system states rather than traditional indirect constraints via error states. By employing a dynamic reconstruction method based on auxiliary barrier function, the position-constrained system is transformed into an unconstrained one, enabling synchronous handling of both constrained and unconstrained systems. Combined with the designed non-singular terminal integral sliding mode and non-integer power structure, this framework ensures precise bilateral synchronization characteristics and finite-time convergence performance. Neural networks and adaptive learning strategies are adopted to address system dynamic uncertainties, while stability analysis and finite-time convergence proofs are completed using Lyapunov methods. Experimental results demonstrate that the proposed control method effectively maintains outputs within preset boundaries while coping with time-varying delays and model uncertainties, exhibiting outstanding control performance.