Activation cues and force scaling methods for virtual fixtures

Previous work has shown that virtual fixtures applied to cooperative manipulation systems improve performance in tracking tasks. However, if a user wishes to move away from the nominal path, the stiffness of the virtual fixture should be adjusted to increase the level of user control. Our experiments show that modifying the stiffness of a soft virtual fixture can improve a user's performance, measured by position error and execution time, for a task where the user targets a point off the nominal path. Three methods of changing the stiffness of the virtual fixture were tested: on/off toggle, linear fade, and constant force hold. In addition, we considered three methods of activating the stiffness change: explicit, implicit, and automatic. User trials revealed that several combinations of stiffness scaling and activation methods significantly reduced position errors when compared to no stiffness adjustment.