Learning-Based Planning for Improving Science Return of Earth Observation Satellites

2025-09-05

AI Summary

Earth observation satellites face fundamental limitations—including orbital constraints, narrow field-of-view, and high energy costs for attitude maneuvering—that hinder efficient acquisition of high-information-content data. To address this, we propose a dynamic pointing framework that leverages predictive instrument data and intelligent mission planning to enable adaptive, real-time optimization of the primary sensor’s pointing direction. Our approach innovatively integrates reinforcement learning with imitation learning to construct a resource-aware decision-making model, enabling effective training under low-data regimes (i.e., small-sample settings). Experimental results demonstrate that the two learning-based strategies improve sampling performance by 13.7% and 10.0%, respectively, over conventional heuristic methods. Simulation studies confirm superior scientific data acquisition efficiency and higher information density, while also validating robustness under limited training data—indicating strong potential for on-orbit deployment.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: State Estimation
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Earth observing satellites are powerful tools for collecting scientific information about our planet, however they have limitations: they cannot easily deviate from their orbital trajectories, their sensors have a limited field of view, and pointing and operating these sensors can take a large amount of the spacecraft's resources. It is important for these satellites to optimize the data they collect and include only the most important or informative measurements. Dynamic targeting is an emerging concept in which satellite resources and data from a lookahead instrument are used to intelligently reconfigure and point a primary instrument. Simulation studies have shown that dynamic targeting increases the amount of scientific information gathered versus conventional sampling strategies. In this work, we present two different learning-based approaches to dynamic targeting, using reinforcement and imitation learning, respectively. These learning methods build on a dynamic programming solution to plan a sequence of sampling locations. We evaluate our approaches against existing heuristic methods for dynamic targeting, showing the benefits of using learning for this application. Imitation learning performs on average 10.0% better than the best heuristic method, while reinforcement learning performs on average 13.7% better. We also show that both learning methods can be trained effectively with relatively small amounts of data.

Problem

• Optimizing satellite data collection with limited resources
• Improving science return via learning-based dynamic targeting
• Enhancing satellite sensor planning using reinforcement and imitation learning

Innovation

• Learning-based planning for satellite observation optimization
• Reinforcement and imitation learning for dynamic targeting
• Dynamic programming to plan sampling location sequences