Automated loading of a hydraulic excavator using nonlinear model predictive control with preference-based calibration

This study deals with the control methods for automating operation of a hydraulic excavator loading soil onto bed dump truck. The requires moving bucket to target position without contacting In addition, there is demand minimize spillage during process. We applied Model Predictive Control (MPC) using constraints and objective functions corresponding requirements problem. To achieve desired behaviour MPC, weights function need be tuned appropriately. Tuning these not an easy task, even engineers. solve this problem, we utilize weight tuning method based on preference learning. Using method, can by repeating process selecting result that user finds preferable from pairwise results. confirm effectiveness proposed experiments were conducted in simulation environment multiple users different levels skill knowledge. Through experiments, confirmed each was able tune appropriately reflected their preferences, independent