A deep motion reliability scheme for robotic operations

Abstract

This paper proposes a motion reliability analysis method for robotic operations considering uncertainties and external disturbances in the system. First, the kinematic and dynamic model of the robot is established, and the interval number is adopted to describe the uncertain parameters. Then, an observed-based neural adaptive control scheme is developed to guarantee uniform ultimate boundedness of all the signals in the closed loop. Furthermore, a motion reliability model is presented based on interval set operation with the time series of tracking error boundaries. In addition, to improve the computational efficiency, a non-linear autoregressive network with exogenous inputs (NARX) and a segmented sampling technique are proposed to model the dynamic response of the system. Simulation results demonstrate that the proposed analysis method has high accuracy and efficiency in assessing motion reliability.