Automatic Truss Design with Reinforcement Learning

TL;DR

This paper introduces AutoTruss, a two-stage reinforcement learning framework that efficiently generates lightweight, valid truss layouts by combining Monte Carlo tree search and RL, outperforming previous methods in 3D design tasks.

Contribution

AutoTruss is the first effective deep-RL-based approach for truss layout design, combining search and refinement to handle complex physical constraints.

Findings

AutoTruss outperforms previous layouts by 25.1% in 3D cases.

The two-stage framework effectively finds valid and lightweight truss designs.

AutoTruss demonstrates success in both 2D and 3D truss layout problems.

Abstract

Truss layout design, namely finding a lightweight truss layout satisfying all the physical constraints, is a fundamental problem in the building industry. Generating the optimal layout is a challenging combinatorial optimization problem, which can be extremely expensive to solve by exhaustive search. Directly applying end-to-end reinforcement learning (RL) methods to truss layout design is infeasible either, since only a tiny portion of the entire layout space is valid under the physical constraints, leading to particularly sparse rewards for RL training. In this paper, we develop AutoTruss, a two-stage framework to efficiently generate both lightweight and valid truss layouts. AutoTruss first adopts Monte Carlo tree search to discover a diverse collection of valid layouts. Then RL is applied to iteratively refine the valid solutions. We conduct experiments and ablation studies in popular truss layout design test cases in both 2D and 3D settings. AutoTruss outperforms the best-reported layouts by 25.1% in the most challenging 3D test cases, resulting in the first effective deep-RL-based approach in the truss layout design literature.