Simulation-Aided Policy Tuning for Black-Box Robot Learning
This work addresses data-efficient robot learning for robotics applications, presenting an incremental improvement by integrating simulation into policy tuning.
The paper tackles the problem of enabling robots to learn new tasks with minimal data by introducing a black-box policy search algorithm that uses simulation as an additional information source to reduce robot interaction time, demonstrating data-efficient learning in simulated and real robot experiments.
How can robots learn and adapt to new tasks and situations with little data? Systematic exploration and simulation are crucial tools for efficient robot learning. We present a novel black-box policy search algorithm focused on data-efficient policy improvements. The algorithm learns directly on the robot and treats simulation as an additional information source to speed up the learning process. At the core of the algorithm, a probabilistic model learns the dependence of the policy parameters and the robot learning objective not only by performing experiments on the robot, but also by leveraging data from a simulator. This substantially reduces interaction time with the robot. Using this model, we can guarantee improvements with high probability for each policy update, thereby facilitating fast, goal-oriented learning. We evaluate our algorithm on simulated fine-tuning tasks and demonstrate the data-efficiency of the proposed dual-information source optimization algorithm. In a real robot learning experiment, we show fast and successful task learning on a robot manipulator with the aid of an imperfect simulator.