Neural Field Dynamics Model for Granular Object Piles Manipulation
This work addresses the challenge of manipulating granular object piles, which is important for robotics applications in manufacturing and logistics, though it appears to be an incremental advancement in dynamics modeling.
The paper tackles the problem of modeling granular material manipulation by introducing a learning-based dynamics model that uses a density field representation and convolutional neural networks, achieving significant improvements in accuracy and computational efficiency over existing methods while demonstrating zero-shot generalization across various environments.
We present a learning-based dynamics model for granular material manipulation. Inspired by the Eulerian approach commonly used in fluid dynamics, our method adopts a fully convolutional neural network that operates on a density field-based representation of object piles and pushers, allowing it to exploit the spatial locality of inter-object interactions as well as the translation equivariance through convolution operations. Furthermore, our differentiable action rendering module makes the model fully differentiable and can be directly integrated with a gradient-based trajectory optimization algorithm. We evaluate our model with a wide array of piles manipulation tasks both in simulation and real-world experiments and demonstrate that it significantly exceeds existing latent or particle-based methods in both accuracy and computation efficiency, and exhibits zero-shot generalization capabilities across various environments and tasks.