Adaptive Neural Force Control for Micro-Manipulation Subject to Time-Varying Output Constraints and Input Saturation

The compliant-contact interaction in micro-manipulation devices poses significant challenges for force and position control due to the environment’s nonlinear, viscoelastic, and inhomogeneous properties. This brief propose a micro-electro-mechanical manipulation (MMEM)-driven robotic force tracking strategy based on the Kelvin-Boltzmann model subject to unknown dynamics and output constraints. The proposed control utilizes a novel barrier Lyapunov function (BLF) to ensure the output error constraints within reasonable range. First, we constructed a micro-manipulation robot interaction force model based on the Kelvin-Boltzmann framework, which fully considers the actuator saturation constraints of micro-manipulations. Then, considering the output force error constraints in the force-controlled interaction process of micro-manipulation systems, a neural network based BLF is employed in the process of control design to avert the contravention of the time-varying output constraints. Compared to traditional BLF, the designed BLF control strategy can achieve time-varying output constraint control for arbitrary initial conditions. In addition, the proposed neural network controller can effectively addresses actuator saturation constraints. Stability of the system is guaranteed through Lyapunov stability analysis, ensuring convergence of the system. Simulation results validate the feasibility and effectiveness of the proposed NN control strategy, showcasing its superiority.