Adaptive neural network-based fixed-time control for robots with input saturation and prescribed performance

Robotic systems need to meet tracking performance requirements during the execution of specific tasks. However, due to factors such as initial errors or limitations in software and hardware, robotic systems often encounter input saturation constraints. To address these challenges, this paper proposes a fixed-time tracking control algorithm for uncertain robotic systems with prescribed performance and input saturation constraints. Considering the performance requirements of the robotic systems, a controller based on the barrier Lyapunov function is designed. Additionally, an adaptive neural network with fixed-time convergence is developed to estimate the lumped disturbances in the system. To handle potential input saturation constraints, an adaptive auxiliary system is constructed. Based on this, an adaptive neural network-based backstepping anti-saturation prescribed performance controller is designed, enabling the uncertain robot to achieve fixed-time convergence and prescribed performance under the constraints of input saturation. Experiments conducted on the ROKAE collaborative robot validate the practicability and effectiveness of the proposed algorithm.