Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

Holistic Construction Automation with Modular Robots: From High-Level Task Specification to Execution

Created by
  • Haebom

Author

Jonathan K ulz, Michael Terzer, Marco Magri, Andrea Giusti, Matthias Althoff

Outline

This paper proposes a holistic framework to address the problem of robotic automation in construction sites, which is hampered by the constantly changing environment of construction sites, the lack of robotics experts, and the lack of a standardized framework linking robotics and construction practices. Using mobile modular reconfigurable robots, we present a framework for construction task specification, robot shape optimization, and task execution, where users can specify and monitor desired robot motions through a graphical interface. Unlike existing monolithic solutions, we integrate BIM to automatically identify new robots for each task. We utilize modular robot components to rapidly adapt robot hardware to the requirements of specific construction tasks. Unlike previous modular robot optimization studies, we consider multiple competing objectives to explicitly model real-world transition challenges such as calibration errors. We optimize the robot for drilling and spray painting through simulations, and demonstrate successful autonomous execution of robot drilling through experimental validation.

Takeaways, Limitations

Takeaways:
Providing a holistic framework for robotic automation in construction sites
Automatically identifying job-specific robots through BIM integration
Flexible and rapid hardware adaptation using modular robots
Multi-competitive objective optimization considering the difficulties of real-world transfers
Successful experimental verification of autonomous drilling operation
Limitations:
Currently, only simulation and experimental verification results for drilling and spray painting operations are presented, and the generalizability to other various construction operations requires further study.
It may not fully reflect the complexity and unpredictability of actual construction sites.
The performance of the framework may depend on the accuracy and completeness of the BIM data.
Further validation of scalability and practicality for large-scale construction projects is needed.
👍