Automated physics-based modeling of construction equipment through data fusion

被引:0
作者
Xu, Liqun [1 ]
Veeramani, Dharmaraj [2 ]
Zhu, Zhenhua [1 ]
机构
[1] Univ Wisconsin Madison, Dept Civil & Environm Engn, 1415 Engn Dr, Madison, WI 53706 USA
[2] Univ Wisconsin Madison, Dept Ind & Syst Engn, 1513 Univ Ave, Madison, WI 53706 USA
关键词
Construction equipment modeling; Data fusion; Physics-based simulation; Isaac sim; SIMULATORS;
D O I
10.1016/j.autcon.2024.105880
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Physics-based simulations are essential for designing autonomous construction equipment, but preparing models is time-consuming, requiring the integration of mechanical and geometric data. Current automatic modeling methods for modular robots are inadequate for construction equipment. This paper explores automating the modeling process by integrating mechanical data into 3D computer-aided design (CAD) models. A template library is developed with hierarchy and joint templates specific for equipment. During model generation, appropriate templates are selected based on the equipment type. Unspecified joint template data is extracted from technical specifications using a large language model (LLM). The 3D CAD model is then converted into a Universal Scene Description (USD) model. Users can adjust the part names and hierarchy within the USD model to align with the hierarchy template, and joint data is automatically integrated, resulting in a simulation-ready model. This method reduces modeling time by over 87 % compared to manual methods, while maintaining accuracy.
引用
收藏
页数:12
相关论文
共 50 条
[1]  
Alvey B., 2021, Simulated Photorealistic Deep Learning Framework and Workflows to Accelerate Computer Vision and Unmanned Aerial Vehicle Research, P3889
[2]  
[Anonymous], 2024, Caterpillar 390F L | 3D CAD Model Library | GrabCAD
[3]  
[Anonymous], 2024, Working with USD Python Libraries | NVIDIA Developer
[4]  
[Anonymous], 2024, Universal Scene Description in Unreal Engine | Unreal Engine 4.27 Documentation
[5]  
[Anonymous], 2024, Excavator Parts Diagram | Company Wrench
[6]  
[Anonymous], 2024, CAD Converter-Omniverse Extensions Latest Documentation
[7]  
[Anonymous], 2024, YouTube
[8]  
[Anonymous], 2024, Models-OpenAI API
[9]  
[Anonymous], 2024, Isaac Sim-Robotics Simulation and Synthetic Data, NVIDIA Developer
[10]  
[Anonymous], 2024, Joints-Physx 5.3.1 Documentation.