TASK AND MOTION PLANNING FOR COLLABORATIVE ROBOTIC CONSTRUCTION WITH DEFORMABLE MATERIALS
Advances in construction automation research are currently validating a shift from industrial machines towards distributed and mobile material-robot construction systems. This entails distributed robotic systems being co-designed directly with the material systems they employ. Much like pre-industrial construction, the potential of these systems lies in their adaptability and robustness, allowing them to operate in dynamic environments, to collaborate in large teams, and at potentially unlimited scales. This research synthesizes AEC and AI to address the problem of task and motion planning for such multi-machine systems. We will investigate how to combine reinforcement learning (RL) with Logic-Geometric Programming (LGP) to form a task and motion planning strategy that can solve collaborative construction problems.
Therefore the scope of this project entails the co-development of: (1) a collaboration-centric material-robot construction system, (2) a planning algorithm for sequencing tasks from construction artefacts, (3) a system identification methodology for learning material representations, (4) the implementation and evaluation of reinforcement learning (RL) and Logic-Geometric Programming (LGP) in task and motion planning for the sequenced assembly tasks.
Prof. Achim Menges
Institute for Computational Design and Construction (ICD), University of Stuttgart
Nicolas Kubail Kalousdian (ICD)
Prof. Marc Toussaint
Learning and Intelligent Systems Lab (LIS Lab), Technische Universität Berlin
Cyber Valley Research Fund (CyVy-RF)