Sim-to-Real Gap

The article presents a new approach to controlling quadcopters for time-optimal flight using end-to-end reinforcement learning (E2E). Traditional quadcopter control methods face difficulties with unmodeled effects. Recent advancements have brought…