Online Robot Reference Trajectory Adaptation for Haptic Identification of Unknown Force Field

In this paper, a novel online reference trajectory adaptation algorithm is developed, which aims to balance the interactive force and the deviation when the robots interact with an unknown environment. The algorithm first estimates the online parameters of the environmental dynamic model using the Lyapunov-based method. The desired trajectory is then derived by obtaining a tradeoff between the cost of strictly tracking the reference trajectory and the cost of tracking deviation. Simulation studies are carried out to verify the validity of the proposed algorithm. Simulations show that the desired trajectory tends to go around the contour of the force field when more weight is placed on the cost of interaction force, which can be used for haptic identification; identifying contours of the force field.