In this contribution two methods to increase the robustness of robot force control are discussed and experimental results on a real system are presented. One way to control end–effector forces of a robot is to use a force control method which manipulates the desired trajectory of a position controlled robot in a cascaded scheme. The benefit of this control scheme is that the end–effector is also able to follow position trajectories in the non–force–controlled directions. For such a direct parallel force/position control the stability depends on the force controller parameters and the contact stiffness, assuming a stable position control. Tuning these parameters in environments with varying contact stiffness is challenging and time consuming. To avoid this, additional feed–back torques are calculated to increase the dynamic of the force controller. The first method is based on the classical feed–forward control of a robotic system which is used in the feed–back loop of the force control. In the second method the acceleration term of the force control law is used to calculate a feed–back torque.
CITATION STYLE
Parzer, H., Gattringer, H., Neubauer, M., Müller, A., & Naderer, R. (2015). On impact behavior of force controlled robots in environments with varying contact stiffness. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9520, pp. 698–705). Springer Verlag. https://doi.org/10.1007/978-3-319-27340-2_86
Mendeley helps you to discover research relevant for your work.