Markerless Visual Vibration Damping of a 3-DOF Flexible Link Robot Arm

This contribution proposes a model free energy based oscillation damping approach applied to a three degree of freedom (3-DOF) robot arm. The arm experiences gravitational forces and exhibits large deflections. A camera mounted in eye-in-hand configuration provides the endpoint oscillation feedback signal computed from the motion of image features extracted from natural texture in the environment. Sparse features are tracked between two consecutive views. Their mean displacement and the reconstructed camera orientation rate of change constitutes the oscillation feedback signal. The feasibility and limits of image based oscillation and motion detection are discussed in detail. The proposed endeffector vibration damping techniques are analyzed in simulations and further validated in realistic experiments.