Project Gallery
Dis-Co - Artificial Reasoning in Human-Robot Collaborative Disassembly Processes
Seamless collaboration between humans and robots is essential yet challenging, particularly in complex tasks like selective disassembly within the Architecture, Engineering, and Construction (AEC) industry. Traditionally, robot control demands specialized expertise from human operators, limiting natural interaction. By integrating advanced technologies, we can enhance robots' roles beyond mere execution to include understanding context, reasoning logically, and adapting to tasks, enabling more intuitive and effective human-robot collaboration.
To this end, this research explores using Large Language Models (LLMs) to improve human-robot collaboration within the AEC industry. By enabling robots to interpret and act on natural language, LLMs facilitate intuitive communication and provide the system a resemblance of reality, making interactions within disassembly more natural and accessible, minimizing the need for worker retraining, and easing technology adoption.
This research emphasizes robots' critical role in understanding the HRC process. By integrating LLM workflows, robots can understand the context of tasks, reason logically, and adapt their actions accordingly. Furthermore, the robot can advise and provide insight that might be missed.
Implementing this approach encourages the broader use of speech-driven machine control in construction, leading to interactions where workers and robots collaborate as naturally as human colleagues. The robot's insightful input empowers workers to make more informed decisions, enhancing safety and efficiency in construction workflows.
ITECH M.Sc. Thesis Project 2024: Dis-Co - Artificial Reasoning in Human-Robot Collaborative Disassembly Processes
Zahra Shakeri, Shirin Shevidi, Samuel Slezák
Thesis Advisers: Gili Ron, Felix Amtsberg
Thesis Supervisor: Prof. Achim Menges
Second Supervisor: Prof. Thomas Wortmann