Average-Reward Maximum Entropy Reinforcement Learning for Underactuated Double Pendulum Tasks
This work addresses control challenges for underactuated systems like double pendulums, but it is incremental as it applies existing RL concepts to specific simulation tasks.
The paper tackled swing-up and stabilization tasks for underactuated double pendulums (acrobot and pendubot) using a model-free RL algorithm, achieving improved performance and robustness scores compared to baseline methods without requiring engineered rewards or models.
This report presents a solution for the swing-up and stabilisation tasks of the acrobot and the pendubot, developed for the AI Olympics competition at IROS 2024. Our approach employs the Average-Reward Entropy Advantage Policy Optimization (AR-EAPO), a model-free reinforcement learning (RL) algorithm that combines average-reward RL and maximum entropy RL. Results demonstrate that our controller achieves improved performance and robustness scores compared to established baseline methods in both the acrobot and pendubot scenarios, without the need for a heavily engineered reward function or system model. The current results are applicable exclusively to the simulation stage setup.