Sim and Real: Better Together
This addresses the challenge of efficient and accurate training for autonomous robotic systems, offering an incremental improvement over existing sim-to-real methods.
The paper tackles the problem of training controllers for robotic manipulation by learning simultaneously from both simulation and real-world interaction, proposing an algorithm that balances high-throughput but less accurate simulation samples with low-throughput, high-fidelity real samples using separate replay buffers. It demonstrates efficacy in a sim-to-real environment, with theoretical convergence analysis provided.
Simulation is used extensively in autonomous systems, particularly in robotic manipulation. By far, the most common approach is to train a controller in simulation, and then use it as an initial starting point for the real system. We demonstrate how to learn simultaneously from both simulation and interaction with the real environment. We propose an algorithm for balancing the large number of samples from the high throughput but less accurate simulation and the low-throughput, high-fidelity and costly samples from the real environment. We achieve that by maintaining a replay buffer for each environment the agent interacts with. We analyze such multi-environment interaction theoretically, and provide convergence properties, through a novel theoretical replay buffer analysis. We demonstrate the efficacy of our method on a sim-to-real environment.