Structural Design Through Reinforcement Learning Article Swipe

MAKE IMPACT

Thomas Rochefort-Beaudoin , Aurélian Vadean , Niels Aage , Sofiane Achiche · YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2407.07288

This paper introduces the Structural Optimization gym (SOgym), a novel open-source Reinforcement Learning (RL) environment designed to advance machine learning in Topology Optimization (TO). SOgym enables RL agents to generate physically viable and structurally robust designs by integrating the physics of TO into the reward function. To enhance scalability, SOgym leverages feature-mapping methods as a mesh-independent interface between the environment and the agent, allowing efficient interaction with the design variables regardless of mesh resolution. Baseline results use a model-free Proximal Policy Optimization agent and a model-based DreamerV3 agent. Three observation space configurations were tested. The TopOpt game-inspired configuration, an interactive educational tool that improves students' intuition in designing structures to minimize compliance under volume constraints, performed best in terms of performance and sample efficiency. The 100M parameter version of DreamerV3 produced structures within 54% of the baseline compliance achieved by traditional optimization methods and a 0% disconnection rate, an improvement over supervised learning approaches that often struggle with disconnected load paths. When comparing the learning rates of the agents to those of engineering students from the TopOpt game experiment, the DreamerV3-100M model shows a learning rate approximately four orders of magnitude lower, an impressive feat for a policy trained from scratch through trial and error. These results suggest RL's potential to solve continuous TO problems and its capacity to explore and learn from diverse design solutions. SOgym provides a platform for developing RL agents for complex structural design challenges and is publicly available to support further research in the field.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2407.07288

PDF

https://arxiv.org/pdf/2407.07288

OA Status

green

Cited By

2

Related Works

10

OpenAlex ID

https://openalex.org/W4400600675

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4400600675

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2407.07288

Digital Object Identifier

Title

Structural Design Through Reinforcement Learning

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2024

Year of publication

Publication date

2024-07-10

Full publication date if available

Authors

Thomas Rochefort-Beaudoin, Aurélian Vadean, Niels Aage, Sofiane Achiche

List of authors in order

Landing page

https://arxiv.org/abs/2407.07288

Publisher landing page

PDF URL

https://arxiv.org/pdf/2407.07288

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2407.07288

Direct OA link when available

Concepts

Reinforcement, Reinforcement learning, Computer science, Psychology, Artificial intelligence, Social psychology

Top concepts (fields/topics) attached by OpenAlex

Cited by

2

Total citation count in OpenAlex

Citations by year (recent)

2025: 1, 2024: 1

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload