A Comparative Study of Different Control Strategies for a Magnetorheological Haptic System

Abstract

Magnetorheological (MR) dampers are a new generation of adjustable dampers that can generate large resistive forces against motion with low power. This feature is a unique characteristic that idealizes them for many applications including haptics. A haptic interface system is developed that includes a magnetorheological (MR) damper on the master side and a linear DC motor on the slave side. In this system, the slave motion follows the master and the force experienced by the slave is reflected to the master. The major challenge in this arrangement is control of the MR damper to create the desired resistive force. This paper investigates the different strategies for controlling the MR damper. In haptics, rather than the perfect tracking of a desired force, it is more convenient to generate a force in the master that is ‘similar’ to the actual force sensed by the slave. In addition, in many applications of MR dampers such as suspension systems, seismic response reduction, etc., perfect tracking is unneeded. Considering all of these factors, this paper proposes the concept of ‘similarity’ between the desired force and the resistive force of the MR damper. ‘Similarity’ means whenever the desired force is large, the resistive force should also be large, and whenever the desired force is small, the resistive force should then be small. Based on this concept, three controllers are applied and tested in both open loop and closed loop schemes: a gain controller, a fuzzy logic controller, and a lookup table controller. The results of these controllers are evaluated and compared to each other with respect to smaller time delays and smaller and smoother control inputs. In the absence of major differences in the results, the closed loop gain controller appeared to be a suitable choice for control because of its simple scheme and its ease of application.