Knowledge Graph-Based Multi-Agent Path Planning in Dynamic Environments using WAITR

TL;DR

This work tackles multi-agent path planning for data collection in dynamic environments by introducing WAITR, a framework that fuses a dynamic knowledge graph with pathlet-based planning and a cumulative inter-temporal reward objective. It integrates a PREP Mapper for prioritized POI clustering and a TED Predictor for temporal POI shifts, optimizing trajectories with $ \max_{P \subseteq V} \sum_{t=0}^{T} \gamma^{t} \bigg( \sum W(P_{t,i}) - \lambda R(\cdot) \bigg) $. Experimental results in the Gulf of Mexico show WAITR achieving up to 27.1% greater event coverage than a greedy baseline, with improved hazard avoidance and scalable multi-agent coordination via pathlets and knowledge graph updates. The approach demonstrates significant potential to enhance long-term data coverage and navigational safety in dynamic domains and can be extended to other autonomous sensing applications. By combining real-time adaptation with foresight-influenced planning, WAITR offers a robust, domain-agnostic framework for intelligent, risk-aware data collection in complex environments.

Abstract

This paper addresses the challenge of multi-agent path planning for efficient data collection in dynamic, uncertain environments, exemplified by autonomous underwater vehicles (AUVs) navigating the Gulf of Mexico. Traditional greedy algorithms, though computationally efficient, often fall short in long-term planning due to their short-sighted nature, missing crucial data collection opportunities and increasing exposure to hazards. To address these limitations, we introduce WAITR (Weighted Aggregate Inter-Temporal Reward), a novel path-planning framework that integrates a knowledge graph with pathlet-based planning, segmenting the environment into dynamic, speed-adjusted sub-regions (pathlets). This structure enables coordinated, adaptive planning, as agents can operate within time-bound regions while dynamically responding to environmental changes. WAITR's cumulative scoring mechanism balances immediate data collection with long-term optimization of Points of Interest (POIs), ensuring safer navigation and comprehensive data coverage. Experimental results show that WAITR substantially improves POI coverage and reduces exposure to hazards, achieving up to 27.1\% greater event coverage than traditional greedy methods.

Figures (8)

• Figure 1: ROBUSTnet, Aggregated Spatiotemporal Graph
• Figure 2: Flowchart of the data processing pipeline incorporating the Knowledge Graph, Pathlet Selection, and Scoring mechanisms in the WAITR Algorithm.
• Figure 3: Weighted observation radius with hazards factored in. The score for the observation centroid (151.00) highlights the balance between potential data collection and environmental risks.
• Figure 4: Shows densest cluster identifications for frame 1
• Figure 5: Temporal POI cluster realizations across each timeframe.