Safe Multi-agent Autonomous Planning & Scheduling (SafeMAPS)

Due to the existence of lunar ice in permanently shadowed regions of the Moon’s South pole multiple universities, private companies, NASA, and other government space agencies have designed robotic platforms, mission concepts, science experiments, and sensors specifically for the exploration of this location of the Moon. Some mission concepts propose using a team of low-cost rovers for efficient and resilient exploration. Most ongoing studies related to robotic operations at the lunar poles propose solutions to navigation, trajectory planning, communications, and power management with the assumption that vehicle and operational conditions are deterministic in nature. However, exploring the permanently shaded regions can involve extended durations without access to sunlight for solar charging. Adding to the complexity of this problem, a rover will return to sunlight to recharge under uncertain and extreme environments in the form of unknown lunar terrains. Operation at the lunar poles presents communication limitations caused by low elevation angles between the robotic platform and an earth ground station that can preclude human intervention and demands a system that can identify and respond to anomalies autonomously. Two necessary components for remote exploration of an unknown environment are the high-level autonomous Decision-Making (DM) and fault identification and response algorithms. This study will advance the current state-of-the-art of DM algorithms for multiagent autonomous fault-tolerant systems by leveraging the expertise at NASA Ames in the area of intelligent and adaptive systems to create the Safe Multiagent Autonomous Planning and Scheduling (SafeMAPS) framework.

An advancement to the state-of-the-art in swarm intelligence for space based robotics. Deep Reinforcement Learning (RL) has recently shown promising results for the decentralized multiagent problem. This work will extend previous work by studying alternative learning schemes, integrating SHM & DM, and design a framework suitable for space based computing resources.

This is a mature technology (TRL 7+) — the realistic path in is usually NASA's Technology Transfer Program: licensing an existing NASA patent, or a Space Act Agreement to use NASA facilities/expertise directly. NASA also runs a startup licensing program with no upfront fee for companies formed to commercialize a specific NASA technology.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.