N. Hovakimyan

Publications (2)2.11 Total impact

• Source

  Available from: ntnu.no

  Article: ${cal L}_{1}$ Adaptive Controller for Uncertain Nonlinear Multi-Input Multi-Output Systems With Input Quantization

  Hui Sun, N. Hovakimyan, T. Basar
  [show abstract] [hide abstract]
  ABSTRACT: We study the merits of the adaptive controller for nonlinear uncertain Multi-Input Multi-Output (MIMO) systems in the presence of significant unmodeled dynamics and input quantization. We show that with both logarithmic quantization and uniform quantization, if the quantization is sufficiently dense, the adaptive controller ensures that the output of the MIMO system follows a desired reference signal with uniform transient and steady-state performance bounds. Compared with the same feedback architecture without quantization, the uniform performance bounds have an additional positive additive term, linear in the quantization constant, which can be reduced by increasing the quantization density. Simulations illustrate the performance of the adaptive controller in the presence of quantization.
  IEEE Transactions on Automatic Control 04/2012; · 2.11 Impact Factor
• Conference Proceeding: L1 adaptive controller for systems with input quantization

  Hui Sun, N. Hovakimyan, T. Basar
  [show abstract] [hide abstract]
  ABSTRACT: This paper quantifies performance bounds of L₁ adaptive controllers for linear uncertain systems with input quantization. Two basic types of quantizers are considered: uniform quantizers and logarithmic quantizers. The performance bounds of the L₁ adaptive controller in the presence of input quantization proved to have an additional term, dependent upon the quality of quantization. The signals of the closed-loop L₁ adaptive systems can be rendered arbitrarily close to the corresponding signals of a bounded reference system by increasing the adaptation rate and improving the quantizer. Simulations confirm the theoretical results.
  American Control Conference (ACC), 2010; 08/2010