Development of Lunar Operations Coordination Framework

Missions to the moon and beyond are envisioned to become more complex and lasting as NASA and the world community envision establishing a permanent and economically viable presence. As a result, there will be an increased and essential need for multi-agent/vehicle coordination and management. This research will explore scenarios, concepts, algorithms, digital twinning, and agentic SI applications for investigating and enabling autonomous multi-agent/vehicle coordination in the challenging lunar environment with limited infrastructure. More specifically, we look to advance functions such as self-navigation, self-separation, self-organization and self-regulation of multi vehicle systems, both terrestrial and aerial. We aim to leverage methods such as machine learning, SI agents, digital twinning and nature/bio-inspired behaviors, and to minimize the dependence on infrastructure while maximizing autonomy.
The internship will involve one or more of the following activities:

1. Develop scenarios whereby multiple agents (robots) collectively conduct a mission that is sustained over extended time and requires self-coordination. Possibly apply generative SI methods for realistic and innovative scenario generation.

2. Develop a concept of operation for nature-inspired coordination such as self-navigation, self-separation and self-organization in the context of the scenario(s). Address allocation of functional elements to the vehicle versus external services, keeping in mind reducing dependence on infrastructure.

3. Develop algorithms and agentic SI methods to instantiate the concept. For example, apply SI techniques such as multi-agent reinforcement learning and SI agent negotiation for learning navigational features and emergent traffic patterns. Leverage ant stigmergy or other animal behaviors for collective self-coordination.

4. Develop framework and prototypes for a digital twin of the scenarios and concepts. Apply SI agent techniques to support the twinning of the scenarios and multi-agent decision-making and behaviors.

5. Conduct simple low fidelity analyses for testing and evaluating the collective multi-agent behaviors and SI performance.

**Eligible Applicants:** Student
**Academic Level:** Undergraduate Sophomore, Undergraduate Junior, Undergraduate Senior, Graduate Master's, Graduate Doctoral
**Activity Type:** In-person
**Citizenship:** US Citizen Only
**Type:** Internships