Agent based intelligent decision support technology for Earth observation

Space-based Earth observation systems play a significant role in national economic development and national defense security. To achieve all-weather, near real-time integrated observation services encompassing land, sea, air, and space, satellite mission planning systems are evolving towards intelligence and autonomy. This study constructs three typical space-based Earth observation intelligent agents. The intelligent agent models integrate deep learning and reinforcement learning algorithms, and their performance is validated through simulation using reward functions. This research proposes an intelligent agent-based decision support technology aimed at enhancing the efficiency and effectiveness of space-based Earth observation technology. Experimental results show that the intelligent agent-based decision support technology can effectively integrate multi-source observation data, respond quickly to user needs, provide high-quality decision support, and significantly improve the autonomy and intelligence level of satellite mission planning systems.