Event-Based Adaptive Consensus Control of Multiple Flexible Manipulator Systems With Input-Output Constraints.

Abstract

The article investigates the vibration suppression and bipartite consensus control of networked flexible manipulators with input-output constraints. Neural networks (NNs) are utilized to address system uncertainties. The disturbance-like terms generated by backlash and input quantization decomposition, and the time-varying disturbance, are collectively regarded as disturbance terms, and their upper bounds are estimated using adaptive techniques. The integral barrier Lyapunov function (IBLF) is employed to ensure time-varying constraints on both boundary displacement and angle positions. Since the IBLF imposes constraints directly on the system states rather than on the errors, it relaxes the conservative constraints of the traditional BLF control on the state constraints. In addition, to implement input quantization, a hysteresis quantizer is introduced, and the quantized input is decomposed accordingly. An event-triggered mechanism with a relative threshold strategy is designed to reduce the controller update frequency and save network resources.