Decentralized Tracking Control Scheme for Interconnected Nonlinear Systems with Input Saturation and Unknown Control Coefficient

This paper introduces an innovative decentralized adaptive tracking control strategy designed for interconnected systems by input saturation and unknown control coefficient. Broad learning system (BLS), a novel nonlinear approximation technique, is utilized for the identification of the unknown function. Leveraging the properties of Gaussian function, the auxiliary dynamic system is introduced to compensate for the input saturation effect. The Nussbaum-type function is introduced to counteract the difficulty produced by unknown control coefficient existed in the considered interconnected systems. On this basis, the decentralized control scheme is presented for the interconnected nonlinear systems utilizing both of adaptive control and dynamic surface control (DSC). To address the challenge of the “explosion of complexity”, a linear filter, accompanied by a compensation signal, is introduced. According to Lyapunov theory, all signals within the closed-loop systems remain bounded and the tracking errors exhibit convergence to the compact set near the origin. Ultimately, simulation studies are conducted to demonstrate the practicality of the approach.