Study on the transient response to the point-to-point motion controls on a dual-axes air-bearing planar stage

This note addresses the transient behavior of the different point-to-point motion control strategies. The transition between the high-speed velocity control and the high-precision position feedback often leads to undesirable overshoot and residual oscillation. This study first introduces the use of an integrated H-infinity-chain scattering description (CSD) synthesized controller for a unified point-to-point motion to a high-precision positioning control without switching. This study then compares the control effects with the s-curve trajectory control and an augmented Luenberger observer (ALO)-based control. The s-curve control is a position-dependent implementation and has to be converted into a time-domain trajectory for practical implementation. It is desirable to examine if the trajectory error would accumulate into a large overshoot. The ALO uses an auxiliary system to suppress the thrust ripple but is slower compared with the acceleration-based H-infinity-CSD and s-curve control. It is interesting to see that the H-infinity-CSD controller automatically reaches a similar performance as the s-curve trajectory. Both the s-curve and the H-infinity-CSD control algorithms achieve access time in the range of 200 ms for a 10-mm travel. It is also interesting to notice that the time-domain error for the s-curve trajectory does not accumulate, and the overshoot is < 0.08%. As a comparison, the access time for the integrated H-infinity-CSD controller is 195 ms with an overshoot of 0.097%, and the access time for the s-curve trajectory control is 212 ms with an overshoot of 0.016%.