Speed Estimation Improvement Using Kalman Filter for a Haptic Interface

A haptic system is an articulated mechanical structure with motors, position sensors as well as embedded electronics allowing force feedback. It enables the user to interact with an image in virtual reality through the sense of touch and sight. Our test bench is a single degree of freedom system, which is part of six degrees of freedom Surgicobot surgical robot. The speed of such robots is imposed by the user. Thus, the user can freely move the robot handle away from critical regions and when necessary, the robot generates force constraints to prevent unwanted motion. The robot speed is calculated from the position information. Our testbed is a single degree of freedom haptic interface driven by Permanent Magnet Synchronous Motor (PMSM). PMSM offers low rotor inertia, efficient heat dissipation and compact volume and the control algorithm is implemented using the TMS320F2812 fixed point DSP. This paper describes the implementation of a Kalman filter for speed estimation and compares its performances to the Synchronous Constant Elapsed Time (SCET) method. The SCET method estimates the rotor speed by using rotor position samples coming from a FIFO stack so as to caculate the average speed. The filter design criteria for haptic interface are analyzed. Their influence on system stiffness is illustrated. The effects of the control loops sampling period and Kalman filter cut-off frequency on speed estimation quality and the obtained stiffness are discussed. It is shown that the proposed speed estimation method enhances the speed estimation quality and increases the stiffness by 44% compared with the one obtained by using the SCET method.