Proactive Multiaspect Observation Planning for Heterogeneous Spaceborne SAR Constellations: A Graph-Theoretic and Multiobjective Approach

Abstract

Collaborative multiaspect (MA) synthetic aperture radar (SAR) observations are pivotal for enabling advanced Earth observation capabilities, such as 3-D reconstruction, tomographic imaging, and scattering anisotropy analysis. The recent proliferation of global SAR constellations has led to a diverse landscape of in-orbit platforms—characterized by varied orbits, payloads, and agility—which, in turn, creates unprecedented opportunities to acquire geometrically diverse datasets from heterogeneous sources. However, converting these orbital resources into high-value observation products remains a challenge because efficiently identifying geometrically optimal collaboration opportunities requires searching a high-order combinatorial space. To address the limitations of traditional demand-driven scheduling, this article proposes a comprehensive framework for the proactive planning of collaborative MA observations. The framework’s core innovation is a “discovery-then-optimization” strategy that decouples the computationally expensive geometric validation from the initial topological search. First, a graph-based discovery algorithm transforms the problem by efficiently identifying a set of geometrically plausible candidate solutions—the potential collaboration zones (PCZs)—under the adopted PCZ definition and pruning criteria. This high-quality, prefiltered candidate set is then passed to the second stage, where the selection of an optimal mission plan is formulated as a multiobjective optimization problem. We employ the nondominated sorting genetic algorithm II (NSGA-II) to optimize tradeoffs between resource cost, efficiency, and geometric quality. Simulations utilizing commercial constellations (e.g., ICEYE, Capella, and AIRSAT) and a successful real-world experiment validate the framework. These results demonstrate its high efficiency, scalability, and capability to proactively mine high-quality MA opportunities. This research provides a systematic solution to maximize the scientific return of global heterogeneous in-orbit SAR resources.