Adaptive Control Based on Retrospective Cost Optimization

摘要

We present a discrete-time adaptive control law for stabilization, command-following, and disturbance rejectionnthat is effective for systems that are unstable, multi-input/multi-output, and/or non-minimum phase. The adaptivencontrol algorithm includes guidelines concerning the modeling information needed for implementation. Thisninformation includes the relative degree, the first nonzero Markov parameter, and the non-minimum-phase zeros.nExceptwhen the plant has non-minimum-phase zeroswhose absolute value is less than the plant’s spectral radius, thenrequired zero information can be approximated by a sufficient number of Markov parameters. No additionalninformation about the poles or zeros need be known. We present numerical examples to illustrate the algorithm’sneffectiveness in handling systems with errors in the required modeling data, unknown latency, sensor noise, andnsaturation.