Resilient and Efficient Allocation for Large-Scale Autonomous Fleets via Decentralized Coordination

2025/11/17 by Ashish Kumar Perukari, Perukari, Ashish Kumar, Polina Khoroshevskaya +1
Computer Science · Engineering · #Age of Information Optimization #Computer Science and Game Theory (cs.GT) #FOS: Computer and information sciences #Software System Performance and Reliability #Vehicular Ad Hoc Networks (VANETs)

paper · pdf · doi:10.48550/arxiv.2511.12879

openalex publication_date 2025/11/17 · openalex created_date 2025/11/19 · openalex updated_date 2026/07/28

Abstract

Operating large autonomous fleets demands fast, resilient allocation of scarce resources (such as energy and fuel, charger access and maintenance slots, time windows, and communication bandwidth) under uncertainty. We propose a side-information-aware approach for resource allocation at scale that combines distributional predictions with decentralized coordination. Local side information shapes per-agent risk models for consumption, which are coupled through chance constraints on failures. A lightweight consensus-ADMM routine coordinates agents over a sparse communication graph, enabling near-centralized performance while avoiding single points of failure. We validate the framework on real urban road networks with autonomous vehicles and on a representative satellite constellation, comparing against greedy, no-side-information, and oracle central baselines. Our method reduces failure rates by 30-55% at matched cost and scales to thousands of agents with near-linear runtime, while preserving feasibility with high probability.

Citations

• InfraLib: Enabling Reinforcement Learning and Decision-Making for Large-Scale Infrastructure Management
• Optimizing Agricultural Order Fulfillment Systems: A Hybrid Tree Search Approach
• Efficient Strategy Synthesis for MDPs with Resource Constraints
• Practical Contextual Bandits with Regression Oracles
• Safe Model-based Reinforcement Learning with Stability Guarantees
• OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks
• Risk-Constrained Reinforcement Learning with Percentile Risk Criteria
• Taming the Monster: A Fast and Simple Algorithm for Contextual Bandits
• A comprehensive taxonomy for multi-robot task allocation
• Constrained Consensus and Optimization in Multi-Agent Networks
• Consensus and Cooperation in Networked Multi-Agent Systems
• A Formal Analysis and Taxonomy of Task Allocation in Multi-Robot Systems
• Conditional value-at-risk for general loss distributions

Related

• FUSION: Forecast-Embedded Agent Scheduling with Service Incentive Optimization over Distributed Air-Ground Edge Networks
• Distributed Online Rollout for Multivehicle Routing in Unmapped Environments
• Intelligent Knowledge Distribution: Constrained-Action POMDPs for Resource-Aware Multi-Agent Communication
• A Coalition Game for On-demand Multi-modal 3D Automated Delivery System
• Learning Decentralized Routing Policies via Graph Attention-based Multi-Agent Reinforcement Learning in Lunar Delay-Tolerant Networks