Admittance Control for Robotic Loading: Design and Experiments with a 1-Tonne Loader and a 14-Tonne Load-Haul-Dump Machine

This paper describes the design, tuning, and extensive field testing of an admittance-based autonomous loading controller (ALC) for robotic excavation. Several iterations of the ALC were tuned and tested in fragmented rock pilessimilar to those found in operating minesby using both a robotic 1-tonne capacity Kubota R520S diesel-hydraulic surface loader and a 14-tonne capacity Atlas Copco ST14 underground load-haul-dump (LHD) machine. On the R520S loader, the ALC increased payload by 18% with greater consistency, although with more energy expended and longer dig times when compared with digging at maximum actuator velocity. On the ST14 LHD, the ALC took 61% less time to load 39% more payload when compared to a single manual operator. The manual operator made 28 dig attempts by using three different digging strategies, and had one failed dig. The tuned ALC made 26 dig attempts at 10and11MN target force levels. All 10 11MN digs succeeded while 6 of the 16 10MN digs failed. The results presented in this paper suggest that the admittance-based ALC is more productive and consistent than manual operators, but that care should be taken when detecting entry into the rock pile. (C) 2016 Wiley Periodicals, Inc.