Heavy Metal Acrobatics: Learning Dynamic Control for Skid-Steer Loaders
Abstract
Automation of heavy machinery offers the opportunity to improve productivity in mechanized industries such as construction, mining, and material handling. Yet, automation of such machines is challenged by their complex dynamics, in part because of hydraulic actuation that involves fluids, flexible elements, and nonlinear delays. A concomitant challenge is the size and weight of these machines, which can make experiments on the physical systems exceedingly dangerous. Here, we show that reinforcement learning in simulation, followed by transfer of the learned policies from simulation to reality, can successfully automate acrobatic maneuvers that push a heavy construction machine to the limits of its actuation and have only previously been accomplished by a small number of elite human operators in official world-record performances. We trained policies that perform autonomous acrobatic maneuvers on a John Deere 332G, an operational 5-tonne skid-steer loader. The policies can autonomously swing the machine up onto two wheels and, while balancing on two wheels, drive over rough terrain, trace out specified trajectories, spin around, and swing the bucket back and forth before landing it safely back onto four wheels. This work demonstrates that reinforcement learning and sim-to-real transfer enable dynamic autonomous control of heavy machinery, directly contributing to greater automation in heavy equipment sectors.
Resources
arXiv: 2611.18365
Citation
@article{lorentz2026heavy,
title = {Heavy Metal Acrobatics: Learning Dynamic Control for Skid-Steer Loaders},
author = {Lorentz, Pierre-Jean and Egli, Pascal and Jelavic, Edo and Li, Cliff and
Sundar, Ramitha and Suresh, Harikrishnan and Montgomery, Paul and Redden, Lee and
M{"u}ller, Matthias and Sener, Ozan and Koltun, Vladlen and Hutter, Marco},
journal = {Science Robotics},
volume = {11},
pages = {eaef0894},
year = {2026},
doi = {10.1126/scirobotics.aef0894}
}