The virtual hand. The Pulvertaft Prize Essay for 1996

A Wearable Detector for Simultaneous Finger Joint Motion Measurement

This paper presents a wearable and stretchable detector for simultaneous finger joint motion measurement, providing the feasible technical solutions to solve the problems existing in the current devices. The sensor array installed in the detector consists of 14 custom-made bending sensors and an inertial measurement unit with high accuracy. The glove support for a sensor array, which is made of three silicone materials with different elasticity, not only overcomes the drawbacks of cloth support, but also addresses the compatibility between the stretch substrate and the inelastic sensing components. Besides, two-section electrical connection avoids the problems caused by the external wires in traditional solutions to enhance the durability of glove. The experiment results indicated that the wearable detector offers a good user experience by exerting a small force on fingers. The device also shows a strong durability as it worked properly after tens of thousands of bending. The measurement tests on a single sensor demonstrated that two kinds of sensors are both with high accuracy. Besides, the hand goniometric system has an excellent performance in terms of hand motion detection. The measurement error of the device in subjects with different hand sizes is within a reasonable range. The good properties of the sensory detector allow it to be applied in clinical applications.

A new method of accurate hand- and arm-tracking for small primates

The investigation of grasping movements in cortical motor areas depends heavily on the measurement of hand kinematics. Currently used methods for small primates need either a large number of sensors or provide insufficient accuracy. Here, we present both a novel glove based on electromagnetic tracking sensors that can operate at a rate of 100 Hz and a new modeling method that allows to monitor 27 degrees of freedom (DOF) of the hand and arm using only seven sensors. A rhesus macaque was trained to wear the glove while performing precision and power grips during a delayed grasping task in the dark without noticeable hindrance. During five recording sessions all 27 joint angles and their positions could be tracked reliably. Furthermore, the field generator did not interfere with electrophysiological recordings below 1 kHz and did not affect single-cell separation. Measurements with the glove proved to be accurate during static and dynamic testing (mean absolute error below 2° and 3°, respectively). This makes the glove a suitable solution for characterizing electrophysiological signals with respect to hand grasping and in particular for brain-machine interface applications.