Skip to content
Proceedings › ISARC › 2026

2026 Proceedings of the 43rd ISARC, Singapore

ISBN 978-0-6458322-3-5ISSN 2413-5844332 papersEditors Qian Chen, Gaang Lee, Ci-Jyun Liang, Jiansong Zhang, Vineet Kamat

Papers in this volume

Showing 1–7 of 332

Front matter and table of contents

Qian Chen, Gaang Lee, Ci-Jyun Liang, Jiansong Zhang, Vineet Kamat

Abstract: The International Association for Automation and Robotics in Construction (IAARC) and the Local Organizing Committee are pleased to present the Proceedings of the 43rd International Symposium on Automation and Robotics in Construction (ISARC 2026). This year’s symposium draws inspiration from the vibrant spirit, innovation, and cultural diversity of its host …

Keywords: No keywords

Control of a Robotic Telehandler for Sandwich Panel Assembly: High-Fidelity Digital Twin and Fractional Predictive Controller

Lucas Costa, João Alves, Nuno Ferreira, Micael Couceiro
Pages 1–8

Abstract: On-site sandwich-panel assembly demands safe, precise manipulation by heavy-duty equipment operating under dynamic, partially known constraints. We present motion-control design and validation for a retrofitted rotating telehandler. A ROS2-compatible Unity digital twin is used for safe, repeatable tuning …

Keywords: Construction robotics, Robotic telehandler, ROS, Digital twin, Unity, Fractional-order control, Predictive controller, PID, Sandwich panel assembly, Sim-to-real

Learning Humanoid Loco-Manipulation for Transporting Unwieldy Construction Objects

Zaolin Pan, Yantao Yu, Zhengbo Zou
Pages 9–16

Abstract: Enabling robust loco-manipulation for transporting unwieldy objects is a critical prerequisite for deploying humanoid robots in construction. However, this task is complicated by severe physical constraints, including irregular object geometries, limited actuator torque, and high moments of inertia caused by eccentric centers of mass …

Keywords: Humanoid robots, Loco-manipulation, Construction robotics, Reinforcement learning, Whole-body control

A Plug-and-Play Semantic Execution Framework for Reusable Cross-Domain Automation in Digital Twins

Chien-Pu Huang, Shang-Hsien Hsieh
Pages 17–24

Abstract: This research proposes a plug-and-play semantic execution framework that enables Digital Twin–oriented automation to be configured through reusable domain knowledge rather than project-specific code. The framework is built on Domain Packs—modular semantic units encapsulating ontologies, data mappings, reasoning rules, workflow templates, and provenance policies—and a set of domain-agnostic agents operating …

Keywords: Digital Twin automation, Semantic interoperability, Domain Packs, Multi-agent systems, Knowledge graphs, Built environment

Experimental Validation of a Real-Time Digital Twin for Latency and Performance Analysis in Automated Wood-Framing

Djamel Eddine Touil, Jordan Kolb, Ahmed Bouferguene, Yasser Mohamed, Mohamed Al-Hussein, Simaan AbouRizk
Pages 25–32

Abstract: Digital twins (DTs) have become a cornerstone of cyberphysical systems, enabling real-time monitoring, control, and predictive analytics through continuous data exchange between physical assets and their virtual counterparts. While extensively applied in manufacturing, their adoption in off-site wood construction remains limited, particularly with respect to …

Keywords: Digital Twin, Wood Framing, Construction, Off-Site, BIM

Towards RL Control for Flexible Robotic Applications in Construction: Grasping Static Objects

Mohab Hassaan, Florian Noichl, André Borrmann
Pages 33–40

Abstract: With recent advancements in robotics, robots have been successfully utilized to automate processes in various sectors. However, their use in the construction industry has not mirrored this trend, due to the complexity of construction processes and the fact that they take place on highly dynamic construction sites. Therefore, construction robots …

Keywords: Construction Robotics, Reinforcement Learning, Proximal Policy Optimization, Robotic Grasping

Hybrid Scene Graph Generation Framework for Structural Construction Activity Understanding Leveraging Human-Object Interaction and Vision-Language Models

Bryan D. Tan, Aureley Pradipta, Jacob J. Lin
Pages 41–48

Abstract: Understanding on-site activities is fundamental for supporting construction planning, coordination, and performance assessment; however, this task is commonly performed through manual interpretation by human supervisors, limiting information integration with autonomous systems. Scene graph generation (SGG) provides a structured representation of visual scenes, but existing image-based methods rely heavily on supervised …

Keywords: Scene Understanding, Scene Graph Generation, Human-Object Interaction, Construction Activity Monitoring