Improving agent performance in fluid environments by perceptual pretraining

TL;DR

This paper introduces a perceptual pretraining framework for fluid environments that enables agents to quickly adapt to complex tasks like obstacle detection and vortex avoidance, improving performance through unsupervised learning.

Contribution

The paper presents a novel pretraining framework with an information compression model tailored for fluid environment perception, demonstrated through simulation on a two-cylinder problem.

Findings

Pretrained agents better perceive fluid features and adapt faster.

Unsupervised pretraining enhances multi-scenario task performance.

Sensitivity analysis confirms robustness of the pretrained model.

Abstract

In this paper, we construct a pretraining framework for fluid environment perception, which includes an information compression model and the corresponding pretraining method. We test this framework in a two-cylinder problem through numerical simulation. The results show that after unsupervised pretraining with this framework, the intelligent agent can acquire key features of surrounding fluid environment, thereby adapting more quickly and effectively to subsequent multi-scenario tasks. In our research, these tasks include perceiving the position of the upstream obstacle and actively avoiding shedding vortices in the flow field to achieve drag reduction. Better performance of the pretrained agent is discussed in the sensitivity analysis.