Fast and totally nonlinear set membership filtering; application to control a 5-DOF robot manipulator

Control of a robot manipulator to follow the desired path is a growing need in today industry, especially for manipulators with more than two links which are capable of miscellaneous motions in space. This paper is thoroughly concerned with the issue, taking in mind the necessity of estimation or filtering states from noises. Abstractly saying, the control used is a feedback-feedforward one while the states are estimated by a fast totally nonlinear version of the so-called set membership filtering algorithm.