A new approach to robust adaptive control in the presence of unknown and bounded disturbances

Summary form only given. An adaptive l¹ controller has been based on the certainty-equivalence approach where at each step the system parameters are estimated and used to implement the controller. At each estimation stage a modeling error is committed which affects the output of the estimated plant. The l¹ optimal controller has been constructed to minimize the effect of this error on the output. It has been shown that under reasonable assumptions an adaptive controller can be based on the l¹ design strategy, and the result is a globally convergent adaptive scheme