Stochastic Optimization Methods in Robust Adaptive Control of Robots

Abstract

In optimal control of robots the standard procedure is to determine first off-line an optimal reference trajectory and an optimal open loop control, using some nominal or estimated values of the model parameters, and to correct the resulting deviation of the effective trajectory or performance of the system from the reference trajectory, from prescribed performance values, resp., by online measurement and control actions. However, online measurement and control actions are expensive and time-consuming, moreover, they are suitable only for rather small deviations. By adaptive optimal stochastic trajectory planning and control (AOSTPC), i.e. incorporating sequentially the available prior and sample information about the unknown model parameters into the optimal control design process by using stochastic optimization methods, the conditional mean absolute deviation between the actual and reference trajectory, performance, resp., can be reduced considerably, hence, more robust controls are obtained.