Physics-Aware Combinatorial Assembly Sequence Planning using Data-free Action Masking

2024-08-19

AI Summary

This work addresses the physically executable compositional Assembly Sequence Planning (ASP) problem: given a target object’s 3D geometry, automatically generating a feasible sequence for placing standardized modular units (e.g., LEGO bricks). To overcome the key limitation of conventional methods—frequent physical violations (e.g., floating parts, collisions, structural instability)—we propose a data-free, online, physics-aware action masking mechanism that dynamically filters infeasible actions during reinforcement learning, enabling the first end-to-end ASP policy trained with zero physical violations. Our approach integrates deep reinforcement learning, real-time physics-constrained modeling, 3D structural parsing, and precise collision detection. Evaluated on over 250 complex LEGO assemblies, our method achieves 100% success rate in generating physically valid sequences—significantly outperforming the strongest baseline, which fails on more than 40 instances.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Low Level & Physics-based Vision
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Combinatorial assembly uses standardized unit primitives to build objects that satisfy user specifications. This paper studies assembly sequence planning (ASP) for physical combinatorial assembly. Given the shape of the desired object, the goal is to find a sequence of actions for placing unit primitives to build the target object. In particular, we aim to ensure the planned assembly sequence is physically executable. However, ASP for combinatorial assembly is particularly challenging due to its combinatorial nature. To address the challenge, we employ deep reinforcement learning to learn a construction policy for placing unit primitives sequentially to build the desired object. Specifically, we design an online physics-aware action mask that filters out invalid actions, which effectively guides policy learning and ensures violation-free deployment. In the end, we apply the proposed method to Lego assembly with more than 250 3D structures. The experiment results demonstrate that the proposed method plans physically valid assembly sequences to build all structures, achieving a $100%$ success rate, whereas the best comparable baseline fails more than $40$ structures. Our implementation is available at url{https://github.com/intelligent-control-lab/PhysicsAwareCombinatorialASP}.

Problem

• Ensuring physically executable assembly sequences for combinatorial objects
• Addressing combinatorial challenges in assembly sequence planning
• Filtering invalid actions using physics-aware masking for valid assembly

Innovation

• Deep reinforcement learning for assembly planning
• Physics-aware action mask filters invalid actions
• Ensures 100% success in Lego structure assembly