Research of Dataglove Calibration Method Based on Genetic Algorithms

Dataglove is an important interface device for dexterous interactions with virtual environments, which should be calibrated before used. Calibration is the initialization of dataglove and establishes a mapping from the sensor values to the joint angles of an articulated hand. For the exoskeleton dataglove, we build a measurement model of the finger flexion angle for human hand based on the theory of four-bar-linkage motion stabilization. The four postures obtained by the standard block are used to calibrate the parameters of the dataglove that are solved by the genetic algorithms (GA). The calibration based on GA has competitive merits, namely, insusceptibility to getting trapped in local extremum and no need of initial values. Experiment verifies that the method is simple and the result is accurate