Adaptive fixed-time consensus control for a class of non-strict feedback multi-agent systems subject to input nonlinearities, state constraints, unknown control directions, and actuator faults

• Published in: European journal of control Vol. 66; p. 100649
• Main Authors: Mohit, Mohammaderfan, Shahrokhi, Mohammad
• Format: Journal Article
• Language: English
• Published: Philadelphia Elsevier Ltd 01.07.2022; Elsevier Limited
• Subjects: Actuator fault; Actuators; Adaptive control; Closed loops; Constraints; Control systems design; Controllers; Design; Dynamic stability; Faults; Feasibility studies; Feedback; Fixed-time control; Fuzzy logic; Fuzzy systems; Input nonlinearity; Liapunov functions; Multi-agent systems; Multiagent systems; Neural networks; Non-strict feedback systems; Nonlinearity; State constraints; Tracking errors; Input nonlinearity; Fixed-time control; Non-strict feedback systems; Actuator fault; State constraints; Multi-agent systems
• ISSN: 0947-3580; 1435-5671
• DOI: 10.1016/j.ejcon.2022.100649

In this work, an adaptive fixed-time controller has been designed for a class of uncertain non-strict feedback multi-agent systems (MASs) subject to different types of input nonlinearities, time-varying asymmetric state constraints, unknown control directions, infinite number of actuator faults and external disturbances. Compared to the existing consensus control schemes, the proposed controller can handle input nonlinearities, actuator faults and unknown control directions simultaneously, while guaranteeing fixed-time convergence of the consensus tracking error and satisfying state constraints for non-strict feedback MASs. By employing a nonlinear mapping, the constrained MAS has been transformed into an unconstrained one, hence the feasibility analysis required for barrier Lyapunov function (BLF)-based controllers for state-constrained systems has been avoided. Different input nonlinearities including backlash, hysteresis, dead-zone and saturation have been tackled by employing a unified framework, hence each agent can have a different type of input nonlinearity. By using fuzzy logic systems (FLSs), system unknown dynamics have been approximated. Utilizing Lyapunov stability theorem, it has been proved that all closed-loop signals are semi-globally practically fixed-time stable (SPFTS), state constraints are satisfied and the consensus tracking is achieved. Finally, the effectiveness of the designed control scheme has been shown via two simulation studies.