Reproduce “Adaptive Bi-Level Multi-Robot Task Allocation and Learning under Uncertainty with Temporal Logic Constraints” 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
Xiaoshan Lin, Roberto Tron
An adaptive two-level framework coordinates multi-robot assignments for Time Window Temporal Logic tasks when transition dynamics are only partially known. It combines conservative probability bounds with Wilson-score estimates learned from execution outcomes, preserving task-satisfaction requirements while reallocating excess capacity to auxiliary goals such as monitoring or energy-aware behavior; simulations with eight robots show satisfaction above target thresholds and allocation times below roughly 0.4 seconds for teams of up to 48 robots.
Key insights
- Separating allocation from execution allows the coordinator to reason about task reliability while robots make local decisions, reducing the need for continuous centralized control.
- Static probability bounds provide conservative guarantees but can waste capacity; data-driven bounds adapt to observed performance and allow more robots to pursue auxiliary rewards as uncertainty estimates improve.
- The formal guarantee is conditional on feasible allocation, valid uncertainty bounds, and independence assumptions; it does not cover interacting robots, correlated failures, or collision avoidance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.