DiffPhyCon: A Generative Approach to Control Complex Physical Systems

TL;DR

DiffPhyCon is a novel generative method for controlling complex physical systems that effectively explores control sequences, outperforms existing methods, and reveals insights into fluid dynamics patterns.

Contribution

We introduce DiffPhyCon, a new generative approach that minimizes energy and control objectives simultaneously, enabling global exploration and near-optimal control sequence planning.

Findings

Outperforms classical and deep learning control methods.

Successfully controls complex systems like Burgers' equation, jellyfish movement, and smoke.

Reveals fluid dynamics patterns consistent with established research.

Abstract

Controlling the evolution of complex physical systems is a fundamental task across science and engineering. Classical techniques suffer from limited applicability or huge computational costs. On the other hand, recent deep learning and reinforcement learning-based approaches often struggle to optimize long-term control sequences under the constraints of system dynamics. In this work, we introduce Diffusion Physical systems Control (DiffPhyCon), a new class of method to address the physical systems control problem. DiffPhyCon excels by simultaneously minimizing both the learned generative energy function and the predefined control objectives across the entire trajectory and control sequence. Thus, it can explore globally and plan near-optimal control sequences. Moreover, we enhance DiffPhyCon with prior reweighting, enabling the discovery of control sequences that significantly deviate from the training distribution. We test our method on three tasks: 1D Burgers' equation, 2D jellyfish movement control, and 2D high-dimensional smoke control, where our generated jellyfish dataset is released as a benchmark for complex physical system control research. Our method outperforms widely applied classical approaches and state-of-the-art deep learning and reinforcement learning methods. Notably, DiffPhyCon unveils an intriguing fast-close-slow-open pattern observed in the jellyfish, aligning with established findings in the field of fluid dynamics. The project website, jellyfish dataset, and code can be found at https://github.com/AI4Science-WestlakeU/diffphycon.