Estimation-enhanced disturbance-observer-based discrete sliding mode control of constrained dual-arm space robots

High-precision trajectory control of dual-arm space robots remains challenging when the two manipulators firmly grasp a common target. In this case, the resulting closed-chain constraints tightly couple the base, manipulators, and payload, while external disturbances, measurement noise, and actuator saturation jointly degrade closed-loop performance. This paper proposes an estimation-enhanced disturbance-observer-based discrete integral sliding mode control scheme for constrained dual-arm space robots. The key idea is to combine constraint-consistent state reconstruction with disturbance compensation so that robust high-precision control can be achieved under practical sensing and actuation limitations. First, the closed-chain dynamics are reformulated via the embedding technique, yielding a reduced constrained model suitable for controller and observer synthesis. Second, a constrained extended Kalman filter is introduced to reconstruct reliable states from noisy measurements while preserving kinematic consistency. Third, a second-order disturbance observer is designed on the reduced constrained dynamics to estimate lumped disturbances and provide feedforward compensation. On this basis, a discrete integral sliding mode controller with an anti-saturation auxiliary system is developed to improve tracking accuracy, disturbance rejection, and transient behavior under bounded torque inputs. Lyapunov analysis proves that the closed-loop tracking errors are uniformly ultimately bounded under bounded state-estimation, disturbance-observation, and projected saturation residuals. Comparative simulations, Monte Carlo tests, and sensitivity analyses show that the proposed framework achieves smaller tracking errors, faster convergence, and stronger disturbance rejection than benchmark methods.