Decentralized adaptive tracking of interconnected non-affine systems with time delays and quantized inputs

The problem of decentralized adaptive tracking is investigated for a class of uncertain non-affine systems with time-delay interconnections, quantized inputs and external disturbances. By using mean value theorem, Lyapunov–Krasovskii functional method and dynamic surface control (DSC) technique along with minimal-learning-parameters (MLP) algorithm, a decentralized adaptive fuzzy tracking controller is synthesized. Stability analysis subject to the effect of input quantization is conducted and the proposed memoryless local controller can guarantee semiglobal uniform boundedness of all signals in the closed-loop interconnected system. The main advantage of the proposed method is that for each ni-th order pure-feedback non-affine subsystem, only one parameter is needed to be estimated on-line regardless of the number of fuzzy rule bases used. This fact, along with the DSC technique, can circumvent the problems of “computation explosion” and “dimension curse” to almost the greatest extent. Finally, simulation study is provided to demonstrate the effectiveness and performance of the proposed scheme.