USING A PREDICTIVE GRAPHICAL USER INTERFACE TO IMPROVE TOWER CRANE PERFORMANCE

摘要

Operating tower cranes is very challenging because thernpayload significantly lags behind the control input and undergoesrnlarge amplitude swinging. While significant workrnhas been directed at reducing payload swing, little effortrnhas been placed on reducing the time lag. There is goodrnreason for neglecting the time lag; it cannot be eliminated.rnThe time lag is a result of the physical limitations of therncrane motors, coupled with the very large masses and inertiasrnof tower cranes and their payloads. Good crane operatorsrnbecome accustomed to the time lag and develop thernskill to start decelerating the crane well before the desiredrnstopping location. This paper presents a control methodrnthat aids the human operator by graphically displaying arnprediction of where the crane will stop. This predictive elementrnis combined with an input-shaping controller to reducernthe payload swing. Experimental results show thatrnthe proposed control system significantly improves towerrncrane performance.