A Novel Composite Observer-Based Adaptive Predefined-Time Control for Robotic Manipulators With Deception Attacks and Full-State Constraints

Abstract

To enhance the physical safety and dynamically adjustable response of constrained robotic manipulators under malicious attacks, this paper presents an improved fuzzy state observer-based predefined-time control design for an n-link robotic manipulator with full-state constraints under malicious deception attacks. This paper proposes a more general deception attack assumption for n-link robotic manipulators by only requiring the attack on system output <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</i><sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> to be bounded, which relaxes the conventional restriction that all state attacks must be parameterized and bounded. By designing an improved fuzzy state observer that integrates an extended state observer, this scheme can not only accurately estimate the unavailable system state under deception attacks but also observe and compensate for composite disturbances in real time. To address state constraints, a nonlinear state-dependent function is introduced to replace the traditional barrier Lyapunov function method, which achieves an unconstrained transformation of the state-constrained system through a direct constraint handling approach. Subsequently, by constructing a nonsingular predefined-time filter to avoid the high computational complexity issue, the closed-loop system achieves the predefined-time stability while the tracking error converges to the residual set of the origin within the predefined-time. Finally, simulation results demonstrate the effectiveness and superiority of the proposed control strategy.