A Multiagent Decision-Making Model for Martian Landing Site Selection

Abstract

Selecting suitable landing sites on Mars remains one of the key challenges due to the complex terrain of Mars and the diverse scientific objectives in Mars exploration missions. Landing site selection is inherently a multiobjective optimization problem that requires balancing scientific value and engineering safety involving careful tradeoffs and strategic decision making. To address this challenge, we proposed a multiagent decision-making model for Martian landing site selection (MARS-MAS), which simulates the decision-making process of various participants in landing site selection. By modeling the interactions among managers, engineers, scientists, and evaluators, the MARS-MAS model integrates expert knowledge with intelligent optimization to effectively identify landing sites. We applied the MARS-MAS model to the ancient shoreline region on the eastern of Chryse Planitia and successfully identified three candidate landing ellipses: within McLaughlin Crater, north of Oyama Crater, and upstream of Mawrth Vallis. These areas not only possess significant geological and potential life detection value, but also show strong consistency with candidate sites from previous exploration missions, further validating the model’s effectiveness. The results highlight the advantages of the MARS-MAS model in providing an efficient, and intelligent solution for autonomous landing site selection on Mars and other extraterrestrial bodies, thereby enhancing decision-making capabilities for future deep space exploration missions.

Keywords

• Engineering safety
• landing site selection
• Mars
• multiobjective optimization
• multiple decision agent