Liu, CCCCLiuTsai, MSMSTsaiLin, MTMTLinTang, PYPYTangMENG-SHIUN TSAI2021-01-272021-01-2720200094-114Xhttps://scholars.lib.ntu.edu.tw/handle/123456789/542336© 2020 Elsevier Ltd The nonlinear phenomenon of friction causes significant tracking errors when a platform reverses its moving direction. These tracking errors can cause quadrant protrusion (QP) for circular motion. To solve this problem, the conventional approach is to inject a compensation signal during the period of velocity reversal to overcome the friction effect. The waveform of this compensation signal has not yet been investigated. In this paper, a novel multi-square-pulse compensation (MSPC) method is proposed based on minimization of QP to determine the optimal MSPC waveform. To implement MSPC, the waveform is first computed using an identified servo-friction model. MSPC signals for the trajectories with different velocities and accelerations (Vel/Acc) are discussed, and the performance is compared with that of conventional quadrant error compensation (QEC). To validate the novel MSPC algorithm, experiments were conducted on an XY table. To further improve the experimental results by considering modeling errors, MSPC incorporating a tuning algorithm, called MSPC_TA, is proposed. The experimental results with different Vel/Acc show that QPs can be reduced by 95% by using the MSPC_TA approach.animationFriction compensation; Quadrant protrusion; Multi-square-pulse compensation with; tuning algorithm; ITERATIVE LEARNING CONTROL; FRICTION COMPENSATION; DISTURBANCE-OBSERVER; MODEL; FEED; IDENTIFICATION; SYSTEMS; TABLE; ERROR; SLIP[SDGs]SDG11journal article10.1016/j.mechmachtheory.2020.1038752-s2.0-85082650236WOS:000535894900006https://api.elsevier.com/content/abstract/scopus_id/85082650236