EquiBot: SIM(3)-Equivariant Diffusion Policy for Generalizable and Data Efficient Learning
This addresses the challenge of data-efficient and generalizable imitation learning for robot manipulation, offering incremental improvements in robustness and applicability to unseen scenarios.
The paper tackles the problem of enabling robots to learn from limited data and generalize across diverse environments by proposing EquiBot, a method combining SIM(3)-equivariant neural networks with diffusion models, which reduces data requirements and improves generalization, achieving results with just 5 minutes of human demonstrations per task in real-world mobile manipulation tasks.
Building effective imitation learning methods that enable robots to learn from limited data and still generalize across diverse real-world environments is a long-standing problem in robot learning. We propose Equibot, a robust, data-efficient, and generalizable approach for robot manipulation task learning. Our approach combines SIM(3)-equivariant neural network architectures with diffusion models. This ensures that our learned policies are invariant to changes in scale, rotation, and translation, enhancing their applicability to unseen environments while retaining the benefits of diffusion-based policy learning such as multi-modality and robustness. We show on a suite of 6 simulation tasks that our proposed method reduces the data requirements and improves generalization to novel scenarios. In the real world, with 10 variations of 6 mobile manipulation tasks, we show that our method can easily generalize to novel objects and scenes after learning from just 5 minutes of human demonstrations in each task.