Backlash acceleration control method

Abstract

A backlash acceleration control method capable of more accurately setting a timing for an initiation for a backlash acceleration correction in a servo system that carries out a feedforward control compared with conventional methods. According to this method, the backlash acceleration correction is initiated at the time when the feedforward amount is reversed from positive to negative or vice versa. Furthermore, this initiation time for the backlash acceleration correction can also be determined according to the point of change of the sign of a speed command modified by the feed forward amount. When the feedforward coefficient becomes close to "1", the positional deviation becomes almost "0", and therefore an actual shift movement comes to follow the shift command without time lag. Accordingly, a point where a shift direction of the shift command is reversed will be dispersed, and so an initiation time for the backlash acceleration correction cannot be determined accurately. However, a point (Ts), where a sign of the feedforward amount is reversed, can accurately represent the turning point of the shift direction.

Dependencies are inferred from the claim text.