Measurements of human force control during a constrained arm motion using a force-actuated joystick

Abstract

1. When interacting with the environment, human arm movements may be prevented in certain directions (i.e., when sliding the hand along a surface) resulting in what is called a "constrained motion." In the directions that the movement is restricted, the subject is instead free to control the forces against the constraint. 2. Control strategies for constrained motion may be characterized by two extreme models. Under the active compliance model, an essentially feedback-based approach, measurements of contact force may be used in real time to modify the motor command and precisely control the forces generated against the constraint. Under the passive compliance model the motion would be executed in a feedforward manner, using an internal model of the constraint geometry. The feedforward model relies on the compliant behavior of the passive mechanical system to maintain contact while avoiding excessive contact forces. 3. Subjects performed a task in which they were required to slide the hand along a rigid surface. This task was performed in a virtual force environment in which contact forces were simulated by a two-dimensional force-actuated joystick. Unknown to the subject, the orientation of the surface constraint was varied from trial to trial, and contact force changes induced by these perturbations were measured. 4. Subjects showed variations in contact force correlated with the direction of the orientation perturbation. "Upward" tilts resulted in higher contact forces, whereas "downward" tilts resulted in lower contact forces. This result is consistent with a feedforward-based control of a passively compliant system. 5. Subject responses did not, however, correspond exactly to the predictions of a static analysis of a passive, feedforward-controlled system. A dynamic analysis reveals a much closer resemblance between a passive, feedforward model and the observed data. Numerical simulations demonstrate that a passive, dynamic system model of the movement captures many more of the salient features observed in the measured human data. 6. We conclude that human subjects execute surface-following motions in a largely feedforward manner, using an a priori model of the surface geometry. The evidence does not suggest that active, real time use of force feedback is used to guide the movement or to control limb impedance. We do not exclude, however, the possibility that the internal model of the constraint is updated at somewhat longer latencies on the basis of proprioceptive information.

Physiological Basis of Limb-Impedance Modulation During Free and Constrained Movements

“…1, as used previously (McIntyre et al 1995). The experimental setup was equipped with a seat, a force-actuated joystick, and a video screen connected to a computer to display the virtual scene and to give instructions to the subject (procedure to be used, test number, etc.).…”

“…The motors provided a maximum feedback force of 25 N. An optical encoder measured shaft position with a resolution of 0.09 mm at the endpoint. The grip was equipped with a two-axis force sensor having a range of Ϯ40 N and a resolution of 0.02 N. This system was able to simulate different types of constraints such as a constant force field or contact with a rigid surface (for a description of the joystick performance, see McIntyre et al 1995).…”

Does the brain use sliding variables for the control of movements?

“…Electric motors connected to each of the two axes of the handle generated forces on the arm of the subject. Motor torques were controlled so that forces exerted would reflect the force F (t) = m · x (t) coming from the spring/damper link (for a complete description of the apparatus see McIntyre et al 1995). Each experimental session was performed with and without this force feedback.…”

Persistence of Motor Adaptation During Constrained, Multi-Joint, Arm Movements

“…There is considerable theoretical and experimental support for the notion that reaching movements are generated by the CNS executing a set of commands programmed in advance of the movements (Bock 1993;Ghez et al 1990;Happee 1993;Hoff and Arbib 1992;McIntyre et al 1995;Scheidt and Rymer 2000). Preprogramming of movements is also called "feedforward" control, in contrast to feedback control, where motor commands are generated by the CNS in response to sensory information as movements proceed.…”