Learn2Assemble with Structured Representations and Search for Robotic Architectural Construction

Niklas Funk, Georgia Chalvatzaki, Boris Belousov, Jan Peters
Proceedings of the 5th Conference on Robot Learning, PMLR 164:1401-1411, 2022.

Abstract

Autonomous robotic assembly requires a well-orchestrated sequence of high-level actions and smooth manipulation executions. Learning to assemble complex 3D structures remains a challenging problem that requires drawing connections between target designs and building blocks, and creating valid assembly sequences considering structural stability and feasibility. To address the combinatorial complexity of the assembly tasks, we propose a multi-head attention graph representation that can be trained with reinforcement learning (RL) to encode the spatial relations and provide meaningful assembly actions. Combining structured representations with model-free RL and Monte-Carlo planning allows agents to operate with various target shapes and building block types. We design a hierarchical control framework that learns to sequence the building blocks to construct arbitrary 3D designs and ensures their feasibility, as we plan the geometric execution with the robot-in-the-loop. We demonstrate the flexibility of the proposed structured representation and our algorithmic solution in a series of simulated 3D assembly tasks with robotic evaluation, which showcases our method’s ability to learn to construct stable structures with a large number of building blocks. Code and videos are available at: https://sites.google.com/view/learn2assemble

Cite this Paper

BibTeX
@InProceedings{pmlr-v164-funk22a, title = {Learn2Assemble with Structured Representations and Search for Robotic Architectural Construction}, author = {Funk, Niklas and Chalvatzaki, Georgia and Belousov, Boris and Peters, Jan}, booktitle = {Proceedings of the 5th Conference on Robot Learning}, pages = {1401--1411}, year = {2022}, editor = {Faust, Aleksandra and Hsu, David and Neumann, Gerhard}, volume = {164}, series = {Proceedings of Machine Learning Research}, month = {08--11 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v164/funk22a/funk22a.pdf}, url = {https://proceedings.mlr.press/v164/funk22a.html}, abstract = {Autonomous robotic assembly requires a well-orchestrated sequence of high-level actions and smooth manipulation executions. Learning to assemble complex 3D structures remains a challenging problem that requires drawing connections between target designs and building blocks, and creating valid assembly sequences considering structural stability and feasibility. To address the combinatorial complexity of the assembly tasks, we propose a multi-head attention graph representation that can be trained with reinforcement learning (RL) to encode the spatial relations and provide meaningful assembly actions. Combining structured representations with model-free RL and Monte-Carlo planning allows agents to operate with various target shapes and building block types. We design a hierarchical control framework that learns to sequence the building blocks to construct arbitrary 3D designs and ensures their feasibility, as we plan the geometric execution with the robot-in-the-loop. We demonstrate the flexibility of the proposed structured representation and our algorithmic solution in a series of simulated 3D assembly tasks with robotic evaluation, which showcases our method’s ability to learn to construct stable structures with a large number of building blocks. Code and videos are available at: https://sites.google.com/view/learn2assemble} }
Endnote
%0 Conference Paper %T Learn2Assemble with Structured Representations and Search for Robotic Architectural Construction %A Niklas Funk %A Georgia Chalvatzaki %A Boris Belousov %A Jan Peters %B Proceedings of the 5th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2022 %E Aleksandra Faust %E David Hsu %E Gerhard Neumann %F pmlr-v164-funk22a %I PMLR %P 1401--1411 %U https://proceedings.mlr.press/v164/funk22a.html %V 164 %X Autonomous robotic assembly requires a well-orchestrated sequence of high-level actions and smooth manipulation executions. Learning to assemble complex 3D structures remains a challenging problem that requires drawing connections between target designs and building blocks, and creating valid assembly sequences considering structural stability and feasibility. To address the combinatorial complexity of the assembly tasks, we propose a multi-head attention graph representation that can be trained with reinforcement learning (RL) to encode the spatial relations and provide meaningful assembly actions. Combining structured representations with model-free RL and Monte-Carlo planning allows agents to operate with various target shapes and building block types. We design a hierarchical control framework that learns to sequence the building blocks to construct arbitrary 3D designs and ensures their feasibility, as we plan the geometric execution with the robot-in-the-loop. We demonstrate the flexibility of the proposed structured representation and our algorithmic solution in a series of simulated 3D assembly tasks with robotic evaluation, which showcases our method’s ability to learn to construct stable structures with a large number of building blocks. Code and videos are available at: https://sites.google.com/view/learn2assemble
APA
Funk, N., Chalvatzaki, G., Belousov, B. & Peters, J.. (2022). Learn2Assemble with Structured Representations and Search for Robotic Architectural Construction. Proceedings of the 5th Conference on Robot Learning, in Proceedings of Machine Learning Research 164:1401-1411 Available from https://proceedings.mlr.press/v164/funk22a.html.

Related Material