Shape matching based real-coded genetic algorithm for grasping construction

Data glove is widely used in the area of human computer interaction (HCI). Resulting from the complicated architecture of human hand and limited precision of peripheral equipment, the errors for sensorpsilas output must be calibrated. A tempting solution to overcome this issue would be to use external sensors or construct manually to provide the ground-truth data to handle the calibration. The drawbacks of these approaches are tedious calibration procedures and lack of automation. The work described in this paper provides a method for grasping construction, which is based on real-coded generic algorithm and shape matching theory. This approach has the advantage that it does not require any special outer device and can be implemented automatically.