Bridging Perception and Action: A Lightweight Multimodal Meta-Planner Framework for Robust Earth Observation Agents

TL;DR

The paper introduces LMMP, a lightweight multimodal meta-planner for Earth Observation agents that improves planning accuracy and robustness by integrating expert knowledge and a dual-awareness mechanism.

Contribution

It presents a novel framework combining multimodal grounding, expert knowledge injection, and a two-stage training pipeline for more reliable EO planning.

Findings

Significantly improves tool-calling accuracy and task success rates.

Enhances performance across diverse EO missions and executor backbones.

Demonstrates robustness in dynamic, real-world EO scenarios.

Abstract

Autonomous Earth Observation (EO) agents are transitioning from passive perception to complex, multi-step task execution. However, current architectures that integrate planning and execution within a single model often struggle with combinatorial complexity and reasoning errors in dynamic EO scenarios. To resolve these challenges, we propose the Lightweight Multimodal Meta-Planner (LMMP) framework. LMMP incorporates a dual-awareness mechanism that grounds strategic plans in both multimodal image features and high-level task semantics. Crucially, we introduce a Meta Task Library to inject remote sensing expert knowledge directly into the workflow, which standardizes domain logic and ensures plans are physically feasible. We further implement a two-stage training pipeline, initializing the Meta-Planner via expert-distilled Supervised Fine-Tuning and refining it through Direct Preference Optimization based on execution feedback. Extensive experiments on a dataset derived from EarthBench and ThinkGeo demonstrate that LMMP significantly improves tool-calling accuracy and task success rates. Moreover, the framework exhibits strong ``plug-and-play'' versatility, consistently enhancing the performance of diverse executor backbones across previously unseen EO missions.