AssemblyBench: Physics-Aware Assembly of Complex Industrial Objects

• URL: http://arxiv.org/abs/2605.12845v1
• Date: Wed, 13 May 2026 00:44:09 GMT
• Title: AssemblyBench: Physics-Aware Assembly of Complex Industrial Objects
• Abstract summary: Existing datasets focus on simplified scenarios, overlooking shape complexities and assembly trajectories in industrial assemblies.<n>We introduce AssemblyBench, a synthetic dataset of 2,789 industrial objects with multimodal instruction manuals, corresponding 3D part models, and part assembly trajectories.<n>We also propose a transformer-based model, AssemblyDyno, which uses the instructional manual and the 3D shape of each part to jointly predict assembly order and part assembly trajectories.
• Score: 36.01972955466693
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Assembling objects from parts requires understanding multimodal instructions, linking them to 3D components, and predicting physically plausible 6-DoF motions for each assembly step. Existing datasets focus on simplified scenarios, overlooking shape complexities and assembly trajectories in industrial assemblies. We introduce AssemblyBench, a synthetic dataset of 2,789 industrial objects with multimodal instruction manuals, corresponding 3D part models, and part assembly trajectories. We also propose a transformer-based model, AssemblyDyno, which uses the instructional manual and the 3D shape of each part to jointly predict assembly order and part assembly trajectories. AssemblyDyno outperforms prior works in both assembly pose estimation and trajectory feasibility, where the latter is evaluated by our physics-based simulations.

Related papers

• PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangement [89.35154754765502]
  PhyScensis is an agent-based framework powered by a physics engine to produce physically plausible scene configurations.<n>Our framework preserves strong controllability over fine-grained textual descriptions and numerical parameters.<n> Experimental results show that our method outperforms prior approaches in scene complexity, visual quality, and physical accuracy.
  arXiv  Detail & Related papers  (2026-02-16T17:55:25Z)
• Assembler: Scalable 3D Part Assembly via Anchor Point Diffusion [39.08891847512135]
  We present Assembler, a scalable and generalizable framework for 3D part assembly.<n>It handles diverse, in-the-wild objects with varying part counts, geometries, and structures.<n>It achieves state-of-the-art performance on PartNet and is the first to demonstrate high-quality assembly for complex, real-world objects.
  arXiv  Detail & Related papers  (2025-06-20T15:25:20Z)
• IAAO: Interactive Affordance Learning for Articulated Objects in 3D Environments [56.85804719947]
  We present IAAO, a framework that builds an explicit 3D model for intelligent agents to gain understanding of articulated objects in their environment through interaction.<n>We first build hierarchical features and label fields for each object state using 3D Gaussian Splatting (3DGS) by distilling mask features and view-consistent labels from multi-view images.<n>We then perform object- and part-level queries on the 3D Gaussian primitives to identify static and articulated elements, estimating global transformations and local articulation parameters along with affordances.
  arXiv  Detail & Related papers  (2025-04-09T12:36:48Z)
• Manual-PA: Learning 3D Part Assembly from Instruction Diagrams [54.555154845137906]
  We present Manual-PA, a transformer-based instruction Manual-guided 3D Part Assembly framework.<n>Our results show that using the diagrams and the order of the parts lead to significant improvements in assembly performance against the state of the art.
  arXiv  Detail & Related papers  (2024-11-27T03:10:29Z)
• TreeSBA: Tree-Transformer for Self-Supervised Sequential Brick Assembly [51.29305265324916]
  We propose a class-agnostic tree-transformer framework to predict the sequential assembly actions from input multi-view images. A major challenge of the sequential brick assembly task is that the step-wise action labels are costly and tedious to obtain in practice. We mitigate this problem by leveraging synthetic-to-real transfer learning.
  arXiv  Detail & Related papers  (2024-07-22T14:05:27Z)
• SPAFormer: Sequential 3D Part Assembly with Transformers [52.980803808373516]
  We introduce SPAFormer, an innovative model designed to overcome the sequential explosion challenge in the 3D Part Assembly (3D-PA) task.<n>As the number of parts increases, the possible assembly combinations increase exponentially, leading to an explosion that severely hinders the efficacy of 3D-PA.<n>Since the sequence of parts conveys construction rules similar to sentences structured through words, our model explores both parallel and autore generation.
  arXiv  Detail & Related papers  (2024-03-09T10:53:11Z)
• Multi-level Reasoning for Robotic Assembly: From Sequence Inference to Contact Selection [74.40109927350856]
  We present the Part Assembly Sequence Transformer (PAST) to infer assembly sequences from a target blueprint. We then use a motion planner and optimization to generate part movements and contacts. Experimental results show that our approach generalizes better than prior methods.
  arXiv  Detail & Related papers  (2023-12-17T00:47:13Z)
• Rearrangement Planning for General Part Assembly [41.0167725449516]
  We present General Part Assembly Transformer (GPAT), a transformer-based model architecture that accurately predicts part poses. Our experiments on both 3D CAD models and real-world scans demonstrate GPAT's generalization abilities to novel and diverse target and part shapes.
  arXiv  Detail & Related papers  (2023-07-01T03:13:17Z)
• 3D Part Assembly Generation with Instance Encoded Transformer [22.330218525999857]
  We propose a multi-layer transformer-based framework that involves geometric and relational reasoning between parts to update the part poses iteratively. We extend our framework to a new task called in-process part assembly. Our method achieves far more than 10% improvements over the current state-of-the-art in multiple metrics on the public PartNet dataset.
  arXiv  Detail & Related papers  (2022-07-05T02:40:57Z)
• Generative 3D Part Assembly via Dynamic Graph Learning [34.108515032411695]
  Part assembly is a challenging yet crucial task in 3D computer vision and robotics. We propose an assembly-oriented dynamic graph learning framework that leverages an iterative graph neural network as a backbone.
  arXiv  Detail & Related papers  (2020-06-14T04:26:42Z)
• Learning 3D Part Assembly from a Single Image [20.175502864488493]
  We introduce a novel problem, single-image-guided 3D part assembly, along with a learningbased solution. We study this problem in the setting of furniture assembly from a given complete set of parts and a single image depicting the entire assembled object.
  arXiv  Detail & Related papers  (2020-03-21T21:19:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.