Bi-directional Domain Adaptation for Sim2Real Transfer of Embodied Navigation Agents
This work significantly reduces the real-world data requirements for training embodied navigation agents, which is a critical problem for robotics researchers and practitioners.
This paper addresses the sim-vs-real gap in embodied navigation by proposing Bi-directional Domain Adaptation (BDA). BDA uses real2sim to bridge the visual domain gap and sim2real to bridge the dynamics domain gap, achieving performance comparable to a policy fine-tuned with 600k samples using only 5k real-world samples, a 120x speed-up.
Deep reinforcement learning models are notoriously data hungry, yet real-world data is expensive and time consuming to obtain. The solution that many have turned to is to use simulation for training before deploying the robot in a real environment. Simulation offers the ability to train large numbers of robots in parallel, and offers an abundance of data. However, no simulation is perfect, and robots trained solely in simulation fail to generalize to the real-world, resulting in a "sim-vs-real gap". How can we overcome the trade-off between the abundance of less accurate, artificial data from simulators and the scarcity of reliable, real-world data? In this paper, we propose Bi-directional Domain Adaptation (BDA), a novel approach to bridge the sim-vs-real gap in both directions -- real2sim to bridge the visual domain gap, and sim2real to bridge the dynamics domain gap. We demonstrate the benefits of BDA on the task of PointGoal Navigation. BDA with only 5k real-world (state, action, next-state) samples matches the performance of a policy fine-tuned with ~600k samples, resulting in a speed-up of ~120x.