抄録
With the development of virtual reality technology, portable force feedback dataglove plays an important role in virtual assembly and telepresence system. Angle measurement is a basic function of force feedback dataglove, in which the measurement accuracy and calibration are often considered as unsolved problems in previous research. For solving these problems, some work is conducted with an exoskeleton force feedback dataglove using pneumatic artificial muscles as actuators. First, a measurement model of the finger flexion angle based on the theory of four-bar-linkage motion stabilization is built. The effect of structure parameter of linkage on the angle measurement and force feedback is analyzed, which has provided a help for the choice of linkage parameters and improved the accuracy of measuring angle. Then a new calibration method, which is called the “four-posture calibration” based on standard block and genetic algorithm, is proposed. Taking index finger as an example, the process of calibration is introduced in detail. Experiment has shown that with the new calibration method more accurate results can be obtained than that in previous work.
