On discrete time error sign identification algorithms with bounded noise

• Published in: [Proceedings 1992] The First IEEE Conference on Control Applications pp. 828 - 834 vol.2
• Main Author: Kennedy, P.J.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1992
• Subjects: Adaptive control; Convergence; Equations; Error correction; Least squares approximation; Least squares methods; MIMO; Programmable control; Signal to noise ratio; Upper bound
• ISBN: 9780780300477; 0780300475
• DOI: 10.1109/CCA.1992.269740

Modified signed error gradient and least-square identification algorithms are given for both single-input-single-output (SISO) and multiple-input-multiple-output (MIMO) systems with defined convergence characteristics. Convergence properties are obtained in the absence and in the presence of bounded noise. The methods presented are applied to a one-step-ahead adaptive controller. The modified discrete-time error sign gradient algorithm (ESGA) and error sign least-square algorithm (ESLSA) are presented. The ESLSA is applied to the implementation of a one-step-ahead adaptive controller. Simulation results are given.< >