Reproduce “Long-Horizon Multi-Robot Rearrangement Planning for Construction Assembly” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Valentin Noah Hartmann, Andreas Orthey, Danny Driess, Ozgur S. Oguz, Marc Toussaint
A scalable multi-robot planning system for complex, long-horizon construction assembly tasks is developed, leveraging problem decomposition and coordination of heterogeneous robot teams. The system demonstrates its effectiveness on large-scale building models and physically executes plans on real robot hardware.
Key insights
- Iterative decomposition into time-embedded subproblems allows the planning system to scale approximately linearly with the number of agents, effectively managing combinatorial complexity.
- Flexible time-embedding and planning to unknown arrival times for actions yield superior robot utilization and lower overall execution times compared to approaches with fixed time steps.
- The 'Keyframes-opt' component, responsible for solving manipulation constraints, consistently constitutes the majority of the computational cost in complex multi-robot planning scenarios.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.