Adaptive Finite-Time Consensus Protocol for Multiple Mechanical Manipulator Under Stochastic Vibration Conditions Based on Command Filtered Backstepping Method

As robotic arms are increasingly mounted on mobile platforms, traditional control methods, which often focus solely on either the robotic arm or the mobile platform, scarcely address the integrated control of the entire system. This often leads to inadequate handling of stochastic vibrations and environmental disturbances. This paper introduces a consensus protocol designed to enhance rapid and precise collaboration among robotic arms on mobile robots. A distributed neural network-based finite-time controller is proposed for a dynamic system model of robotic arms affected by random vibrations. The effectiveness of this protocol in achieving finite-time convergence is substantiated through a Lyapunov-like function. Simulations are presented to demonstrate the efficacy and advantage of the proposed approach, confirming its potential to significantly improve operational performance for complex robotic systems.