AnalysisRoboticsNovember 18, 2026

ETH Zurich RL policy drives 5-ton John Deere loader in wheelies

Read original source →reddit.com

ETH Zurich researchers trained a reinforcement learning policy that lets a 5-ton John Deere loader pop wheelies, balance on two wheels, clear 75 cm obstacles and land safely. The policy transferred from simulation to the real machine with no further training.

1 source

More stories today

Open the live feed