Gramazio Kohler Research
News
Teaching
Research
Projects
Publications
About
Team
Open Positions
Contact
Compas XR
Compas FAB
Impact Printing
Compas Timber
AIXD: AI-eXtended Design
AI-Augmented Architectural Design
AR Timber Assemblies
Architectural Design with Conditional Autoencoders
Integrated 3D Printed Facade
Think Earth SP7
Robotic Plaster Spraying
Additive Manufactured Facade
Human-Machine Collaboration
Timber Assembly with Distributed Architectural Robotics
Eggshell Benches
Eggshell
CantiBox
Autonomous Dry Stone
RIBB3D
Data Driven Acoustic Design
Mesh Mould Prefabrication
Data Science Enabled Acoustic Design
Thin Folded Concrete Structures
FrameForm
Adaptive Detailing
Deep Timber
Robotic Fabrication Simulation for Spatial Structures
Jammed Architectural Structures
RobotSculptor
Digital Ceramics
On-site Robotic Construction
Mesh Mould Metal
Smart Dynamic Casting and Prefabrication
Spatial Timber Assemblies
Robotic Lightweight Structures
Mesh Mould and In situ Fabricator
Complex Timber Structures
Spatial Wire Cutting
Robotic Integral Attachment
Mobile Robotic Tiling
YOUR Software Environment
Aerial Construction
Smart Dynamic Casting
Topology Optimization
Mesh Mould
Acoustic Bricks
TailorCrete
BrickDesign
Echord
FlexBrick
Additive processes
Room acoustics


Deep Timber, ETH Zurich, 2018-2020
Reinforcement Learning for Robotic Assembly of Timber Joints
This research project investigates the application of machine learning to facilitate architectural construction of timber structures using industrial robots. We employ Deep Reinforcement Learning to teach a robot to assemble lap jointed timber beams, i.e. to insert one notched beam into another, based on contact forces sensed by the robot. In comparison with traditional robot programming, the learning approach enables real-time adaptation to inaccuracies and initial misalignment.
Specifically, we use force-torque sensor data and end-effector poses (observations) to learn sequences of robot movements (actions) that lead to a successful completion of the assembly task. Using an adapted policy gradient method, we leverage human demonstrations and train entirely in a physics simulation environment. We show that these trained policies can be deployed on real robots, thus successfully bridging the “reality gap”. We also demonstrate that our policies can generalize to a fair range of situations not encountered in training, for example, to successfully assemble joints which have a slightly different shape, size or initial position. In a broader picture, this translates into an ability of the robotic system to respond to uncertainties and inaccuracies in the real world.


Publications:
Apolinarska, Aleksandra Anna, Matteo Pacher, Hui Li, Nicholas Cote, Rafael Pastrana, Fabio Gramazio, and Matthias Kohler. Robotic Assembly of Timber Joints Using Reinforcement Learning. Automation in Construction 125 (May 2021): 103569. DOI: 10.1016/j.autcon.2021.103569. PDF

Credits:
Gramazio Kohler Research, ETH Zurich

In cooperation with: Autodesk Research, Autodesk Inc., San Francisco (Dr. Hui Li, Nicolas Cote, Erin Bradner)
Collaborators: Dr. Aleksandra Anna Apolinarska, Matteo Pacher, Lukas Stadelmann, Rafael Pastrana, Hannes Mayer


Copyright 2024, Gramazio Kohler Research, ETH Zurich, Switzerland
Gramazio Kohler Research
Chair of Architecture and Digital Fabrication
ETH Zürich HIB E 43
Stefano-Franscini Platz 1 / CH-8093 Zurich

+41 44 633 49 06
Follow us on:
Vimeo | Instagram