Sim-to-real reinforcement learning applied to end-to-end vehicle control

1 code implementation • 14 Dec 2020 • András Kalapos, Csaba Gór, Róbert Moni, István Harmati

Our controller policy is able to control a small-scale robot to follow the right-hand lane of a real two-lane road, while its training was solely carried out in a simulation.

Collision Avoidance reinforcement-learning +2