ResearchBenchmarksControl on Quanser Aero 2 Real worldFollow-55.8Average ReturnCH-PPO-744.904-566.002-387.1-208.198May 21, 2026Evaluation ResultsMethodMethodLinksAverage ReturnCH-PPOMax degree=5, Policy T...Max degree=5, Policy Type=Chebyshev2026.05-55.8CH-ARSMax degree=6, Policy T...Max degree=6, Policy Type=Chebyshev2026.05-164.2PPOPolicy Type=MLP, Sourc...Policy Type=MLP, Source=RL Baselines3 Zoo2026.05-182ARSPolicy Type=MLP, Sourc...Policy Type=MLP, Source=RL Baselines3 Zoo2026.05-718.4