Abstract
This paper describes the design and application of a haptic device to study the mechanical properties of the human arm during interaction with compliant environments. Estimates of the human endpoint admittance can be obtained by recording position deviations as a result of force perturbations. Previous studies attempted to estimate the impedance by recording force as a result of position perturbations, but these experiments do not require a feasible task of human beings. A general problem of force-controlled haptic devices is the occurrence of contact instability, especially where a small virtual mass is required. This negative effect is reduced by the use of a lightweight but stiff manipulator and a robust servo-based admittance controller. The virtual admittance is accurate to at least 13 Hz, attaining a minimum virtual mass of 1.7 kg (isotropic, without damping and stiffness). The properties of known test loads were estimated with an accuracy higher than 98%, up to 20 Hz. The application of the manipulator is evaluated by an experiment with a subject performing a position maintenance task. With this device it is possible to study the adaptability of the neuromuscular system to a variety of environments, enabling a new and functional approach to human motion research.
Original language | English |
---|---|
Pages (from-to) | 151-168 |
Number of pages | 18 |
Journal | Journal of neuroscience methods |
Volume | 129 |
Issue number | 2 |
Early online date | 22 Aug 2003 |
DOIs | |
Publication status | Published - 30 Oct 2003 |
Externally published | Yes |
Keywords
- Arm admittance
- Contact instability
- Haptic device
- Identification
- Reflexive feedback
- Virtual environment