Linear velocity and acceleration estimation of 3 DOF haptic interfaces

Velocity and acceleration of the end effector of a haptic interface are required for haptic rendering in many aspects such as software damping, friction force rendering, and position control, etc. However, due to limited sensor resolution, non-linearity of forward kinematics, high maneuverability of human arm/hand, and high sampling rate requirement, getting a precise and robust velocity and acceleration estimation is very challenging. In this paper, an adaptive 4-state Kalman filter to estimate the velocity and the acceleration of the end effector is proposed considering that the human arm/hand trajectory has at least 5 non-zero derivatives and the skilled movements follows the constrained minimum jerk trajectory planning. The preliminary simulation results show the effectiveness of the proposed method.