Direct adaptive control of force-guided assembly robots using tuned dither

This paper describes a novel method for force guided assembly of complex parts using dither as a perturbation input. In the robotic assembly of parts with poor surface finish, simple force measurement does not provide reliable information for guiding the parts. Due to friction at burrs and irregular surfaces, force signals are very noisy and erratic, preventing reliable sensing and monitoring of the assembly process. In this paper, instead of simply measuring contact forces, the authors take positive actions by actively shaking the end effector and observing the reaction forces to the perturbation in order to obtain rich, reliable information. By taking the correlation between the input perturbation and the resultant reaction forces, the authors can determine the direction of robot motion correctly despite burrs and poor surface finish. First, the principle of force-guided assembly using the dither and correlation technique is described. The control algorithm is formulated as a type of direct adaptive control where the optimal point of a cost function is searched by perturbing the input. The assembly process is then analyzed by using describing functions for modeling the nonlinearity associated with mechanical contacts. Stability conditions and guidelines for tuning the dither and control parameters are obtained. The proposed method is then applied to a practical task, i.e. assembly of sheet metals. Experiments and simulation verify the analytical results as well as demonstrate the usefulness of the method