Saved in:
Bibliographic Details
Main Authors: Xu, Jinghui, Shangguan, Boyi, Zhu, Mengke, Liu, Hao, Jiang, Junhuan, He, Guangjun, Feng, Pengming, Jin, Shichao, Liang, Bin, Chang, Yongzhe, Tan, Junbo, Zhang, Tiantian, Wang, Xueqian
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.04777
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915983909715968
author Xu, Jinghui
Shangguan, Boyi
Zhu, Mengke
Liu, Hao
Jiang, Junhuan
He, Guangjun
Feng, Pengming
Jin, Shichao
Liang, Bin
Chang, Yongzhe
Tan, Junbo
Zhang, Tiantian
Wang, Xueqian
author_facet Xu, Jinghui
Shangguan, Boyi
Zhu, Mengke
Liu, Hao
Jiang, Junhuan
He, Guangjun
Feng, Pengming
Jin, Shichao
Liang, Bin
Chang, Yongzhe
Tan, Junbo
Zhang, Tiantian
Wang, Xueqian
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2605_04777
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bridging Perception and Action: A Lightweight Multimodal Meta-Planner Framework for Robust Earth Observation Agents
Xu, Jinghui
Shangguan, Boyi
Zhu, Mengke
Liu, Hao
Jiang, Junhuan
He, Guangjun
Feng, Pengming
Jin, Shichao
Liang, Bin
Chang, Yongzhe
Tan, Junbo
Zhang, Tiantian
Wang, Xueqian
Multiagent Systems
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.
title Bridging Perception and Action: A Lightweight Multimodal Meta-Planner Framework for Robust Earth Observation Agents
topic Multiagent Systems
url https://arxiv.org/abs/2605.04777