Co-Design Methods for Developing Distributed Cooperative Multi-Robot Systems

Research Project 19-2 (RP 19-2)

CO-DESIGN METHODS FOR DEVELOPING DISTRIBUTED COOPERATIVE MULTI-ROBOT SYSTEMS FOR CONSTRUCTION 

There is a growing interest in distributed robotic systems for construction, as their low cost, mobility and general adaptability promise more effective, cost-efficient and sustainable construction processes and buildings. Compared to existing standards of on-site construction or off-site factories involving heavy-duty industrial machinery, distributed robotic systems involve small, often customised robots assembling building elements directly on site. Although research into multi-robot systems dates back to the early 1980s, distributed robotic systems specifically for construction is still an emerging field that requires the collaboration between researchers in architecture, engineering and construction on the one hand, and artificial intelligence and robotics on the other, for successful development.

Current research and existing approaches to the development of distributed robotic systems for construction are generally derived from either the robotic system or the building system, with early-stage specificity for one of the systems influencing the overall development. In both cases, this can lead to highly constrained systems, resulting in a limited design space and a severely limited range of possible built structures. In the previous Research Project (RP 19-1) we investigated methods for leveraging the building material as part of the robotic kinematic system for parallel construction. We developed a system that combines actuator hardware and building material into a modular robot-material kinematic chain that can be reconfigured throughout the construction process.

While in the previous project essential system parameters, such as the 1DOF kinematics of each robot unit, were predetermined, we now aim to develop more integrated Co-Design methods that overcome current system limitations resulting from overly deterministic, early definition of essential system parameters. The Co-Design approach of a distributed robotic construction project will involve the parallel development of an architectural material system, artefact design methods, robot mechatronic design, and task and motion planning, with all areas continuously influencing each other throughout the entire system development. As a result of the project, we will deploy a team of inexpensive, agile machines to build a three-dimensional architectural structure.

 

PRINCIPAL INVESTIGATORS

Prof. Dr. Metin Sitti
Physical Intelligence Department (MPI-IS, PI), Max Planck Institute for Intelligent Systems, Stuttgart
Prof. Achim Menges
Institute for Computational Design and Construction (ICD), University of Stuttgart
Prof. Dr. Marc Toussaint
Machine Learning and Robotics Lab (IPVS-MLR), University of Stuttgart

TEAM

Dr.-Ing. Tobias Schwinn (ICD)
Nicolas Kubail Kaloudsdian (ICD)
Hyun Gyu Kim (MPI-IS, PI)
Samuel Leder (ICD)
Valentin Noah Hartmann (IPVS-MLR)

 

PEER-REVIEWED PUBLICATIONS

  1. 2024

    1. Leder, S., Kim, H., Sitti, M., & Menges, A. (2024). Enhanced co-design and evaluation of a collective robotic construction system for the assembly of large-scale in-plane timber structures. Automation in Construction, 162, 105390. https://doi.org/10.1016/j.autcon.2024.105390
    2. Leder, S., & Menges, A. (2024). Merging architectural design and robotic planning using interactive agent-based modelling for collective robotic construction. Journal of Computational Design and Engineering, 11(2), Article 2. https://doi.org/10.1093/jcde/qwae028
    3. Zhang, P., Sevim, S., Leder, S., Maierhofer, M., Schwinn, T., & Menges, A. (2024). Multi-Robotic Maypole Braiding. In O. Kontovourkis, M. C. Phocas, & G. Wurzer (Eds.), Data-Driven Intelligence - Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2024) (Vol. 1, pp. 137–146). eCAADe.
  2. 2023

    1. Hartmann, V. N., Ortiz-Haro, J., & Toussaint, M. (2023). Efficient Path Planning In Manipulation Planning Problems by Actively Reusing Validation Effort. Proc. of the Int. Conf. on Intelligent Robots and Systems (IROS). https://arxiv.org/abs/2303.00637
    2. Hartmann, V. N., & Toussaint, M. (2023). Towards computing low-makespan solutions for multi-arm multi-task planning problems.
    3. Leder, S., & Menges, A. (2023). Introducing Agent-Based Modeling Methods for Designing Architectural Structures with Multiple Mobile Robotic Systems. In C. Gengnagel, O. Baverel, G. Betti, M. Popescu, M. R. Thomsen, & J. Wurm (Eds.), Towards Radical Regeneration (pp. 71--83). Springer International Publishing.
    4. Leder, S., & Menges, A. (2023). Architectural design in collective robotic construction. Automation in Construction, 156, 105082. https://doi.org/10.1016/j.autcon.2023.105082
  3. 2022

    1. Kubail Kalousdian, N., Lochnicki, G., Hartmann, V. N., Leder, S., Oguz, O. S., Menges, A., & Toussaint, M. (2022). Learning Robotic Manipulation of Natural Materials with Variable Properties for Construction Tasks. IEEE Robotics and Automation Letters, 1–1. https://doi.org/10.1109/LRA.2022.3159288
    2. Leder, S., Kim, H., Oguz, O. S., Kalousdian, N. K., Hartmann, V. N., Menges, A., Toussaint, M., & Sitti, M. (2022). Leveraging Building Material as Part of the In-Plane Robotic Kinematic System for Collective Construction. Advanced Science, 2201524. https://doi.org/10.1002/advs.202201524
    3. Menges, A., & Wortmann, T. (2022). Synthesising Artificial Intelligence and Physical Performance. Architectural Design, 92(3), Article 3. https://doi.org/10.1002/ad.2819
  4. 2021

    1. Schubert, I., Driess, D., Oguz, O. S., & Toussaint, M. (2021). Learning to Execute: Efficient Learning of Universal Plan-Conditioned Policies in Robotics. NeurIPS 2021 - Neural Information Processing Systems 34.
    2. Toussaint, M., Ha, J.-S., & Oguz, O. S. (2021). Co-Optimizing Robot, Environment, and Tool Design via Joint Manipulation Planning. 2021 IEEE International Conference on Robotics and Automation (ICRA), 6600–6606. https://doi.org/10.1109/ICRA48506.2021.9561256
    3. Łochnicki, G., Kubail Kalousdian, N., Leder, S., Maierhofer, M., Wood, D., & Menges, A. (2021). Co-Designing Material-Robot Construction Behaviors: Teaching distributed robotic systems to leverage active bending for light-touch assembly of bamboo bundle structures. In B. Farahi, B. Bogosian, J. Scott, J. L. García del Castillo y López, K. Dörfler, J. A. Grant, S. Parascho, & V. A. A. Noel (Eds.), Realignments: Toward Critical Computation - ACADIA 2021.
  5. 2020

    1. Hartmann, V. N., Oguz, O. S., Driess, D., Toussaint, M., & Menges, A. (2020). Robust Task and Motion Planning for Long-Horizon Architectural Construction Planning. Proc. of the IEEE Int. Conf. on Intelligent Robots and Systems (IROS).
    2. Leder, S., Kim, H., Oguz, O. S., Hartmann, V., Toussaint, M., Menges, A., & Sitti, M. (2020). Co-Design in Architecture: A Modular Material-Robot Kinematic Construction System. 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) Proceedings of Workshop on Construction and Architecture Robotics.
    3. Leder, S., Kim, H., Ozgur, O. S., Hartmann, V., Toussaint, M., Menges, A., & Sitti, M. (2020). Co-Design in Architecture: A Modular Material-Robot Kinematic Construction System. 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) Proceedings of Workshop on Construction and Architecture Robotics.
    4. Leder, S., Weber, R., Vasey, L., Yablonina, M., & Menges, A. (2020). Voxelcrete: Distributed Voxelized Adaptive Formwork. ECAADe Online 2020 – Architecture and Fabrication in the Cognitive Age Proceedings of the 38th ECAADe Conference 2020. https://papers.cumincad.org/data/works/att/caadria2022_435.pdf
    5. Melenbrink, N., Werfel, J., & Menges, A. (2020). On-site autonomous construction robots: Towards unsupervised building. Automation in Construction, 119, 103312. https://doi.org/10.1016/j.autcon.2020.103312
    6. Nguyen, S. T., Oguz, O. S., Hartmann, V. N., & Toussaint, M. (2020). Self-Supervised Learning of Scene-Graph Representations for Robotic Sequential Manipulation Planning. Proc. of the Annual Conf. on Robot Learning (CORL).
  6. 2019

    1. Leder, S., Weber, R., Bucklin, O., Wood, D., & Menges, A. (2019). Design and prototyping of a single axis, building material integrated, distributed robotic assembly system. 2019 IEEE: 4th International Workshops on Foundations and Applications of Self* Systems (FAS*), 3rd International Workshop on Self-Organised Construction (SOCO).
    2. Leder, S., Weber, R., Wood, D., Bucklin, O., & Menges, A. (2019). Distributed Robotic Timber Construction: Designing of in-situ timber construction system with robot-material collaboration. ACADIA – Ubiquity and Autonomy Proceedings of the ACADIA Conference 2019.
    3. Yablonina, M., & Menges, A. (2019). Distributed Fabrication: Cooperative Making with Larger Groups of Smaller Machines. Architectural Design, 89(2), Article 2. https://doi.org/10.1002/ad.2413

OTHER PUBLICATIONS

  1. 2023

    1. Hartmann, V. N., Ortiz-Haro, J., & Toussaint, M. (2023). Efficient Path Planning In Manipulation Planning Problems by Actively Reusing Validation Effort. Proc. of the Int. Conf. on Intelligent Robots and Systems (IROS). https://doi.org/10.48550/arXiv.2303.00637

DATA SETS

  1. 2024

    1. Leder, S., Siriwardena, L., & Menges, A. (2024). ABxM.DistributedRobotics.RADr: Agent-based Design and Control of multiple Roaming Autonomous Distributed robots (RADr). DaRUS. https://doi.org/10.18419/darus-4058
    2. Leder, S., Kragl, P., & Menges, A. (2024). Roaming Autonomous Distributed robot (RADr). DaRUS. https://doi.org/10.18419/darus-4059
  2. 2022

    1. Nguyen, L., Schwinn, T., Groenewolt, A., Maierhofer, M., Zorn, M. B., Stieler, D., Siriwardena, L., Kannenberg, F., & Menges, A. (2022). ABxM.Core: The Core Libraries of the ABxM Framework. https://doi.org/10.18419/darus-2994
To the top of the page