Education
Society
Infrastructure

Portable, Off-Grid Artificial Intelligence

Your mission:

Create a challenge that, if solved, would incentivize the design of an artificial intelligence (AI) assistant that could operate flawlessly in extreme environments - from the Arctic to Mars - providing critical decision support, equipment diagnostics, and emergency guidance without relying on network power or connectivity.

Focus on defining the problem, not solving it. The solution topic you create will be the focus of student innovation efforts in the next 18 months.

The Ultimate Destination:

Advancing Human Presence Beyond Earth

Understanding why this matters helps you see the bigger picture and focus your topic on challenges that align with NASA’s mission.

Advanced AI systems will be essential partners for humanity’s expansion into space, operating autonomously in extreme environments to manage critical infrastructure, trouble-shoot equipment, and guide astronauts through complex procedures when Earth is hours or days away. 

Think “Cortana” from the Halo video game series. These rugged systems must master operations in isolated, hostile environments while protecting against cosmic hazards, managing resources, and maintaining vital equipment without external support. They will autonomously support both general purpose and task-specific needs such as managing critical systems, like power, environmental and life-support systems, in addition to troubleshooting equipment issues, and helping prioritize tasks during high-stakes missions. By guiding astronauts through complex procedures and analyzing risks in real time, they’ll ensure off-planet operations run smoothly, even when Earth is out of reach for hours or days.

Proving and ultimately mastering these types of AI systems in isolated and hostile environments will demonstrate their ability to support longer, more complex, riskier journeys. For example, initial testing in the arctic could lay the foundation for systems to support initial lunar exploration. And systems that can prove robust enough to survive lunar environmental conditions will establish the blueprints for sustainable deep-space exploration, refining how humans harvest local resources, protect against cosmic hazards, and maintain equipment.

The Flight Plan:

Core Requirements for Mission Success

These top four core requirements for success highlight the key factors your solution topic should address to help us build the foundation while also offering solutions that can benefit us here on Earth today.

Adaptive and self-learning capabilities

Continuously improve through machine learning, adapting to new situations and scenarios, evolving infrastructure profiles and status over time, and updating existing databases. Consider the impact of decisions not only on local performance but on neighboring agents and on the system of systems (the ecosystem).

01

Environmental hardening

Extreme condition operation, radiation and dust protection, pressure management.

03
05

Autonomous decision-making and recommendations for human operations

Broad situational and state awareness, real-time adjustments. Observe, orient, decide, act.

02

Scalability

Be adaptable for use in different contexts that vary in distance and difficulty, including on Earth, in low-Earth orbit, on extended lunar surface missions, and eventually Mars and deep space missions.

04
06

Adaptive and self-learning capabilities

Continuously improve through machine learning, adapting to new situations and scenarios, evolving infrastructure profiles and status over time, and updating existing databases. Consider the impact of decisions not only on local performance but on neighboring agents and on the system of systems (the ecosystem).

01

Autonomous decision-making and recommendations for human operations

Broad situational and state awareness, real-time adjustments. Observe, orient, decide, act.

02

Environmental hardening

Extreme condition operation, radiation and dust protection, pressure management.

03

Scalability

Be adaptable for use in different contexts that vary in distance and difficulty, including on Earth, in low-Earth orbit, on extended lunar surface missions, and eventually Mars and deep space missions.

04
05
06

Ground-Level Relevance:

Driving Change for Earth, First

How does your topic create meaningful change? The most compelling solution topics bridge the needs of Earth and the demands of space, offering scalable, impactful answers to humanity's biggest challenges. Before diving into feasibility, consider how your topic can shape the world today while paving the way for tomorrow.

Can it scale?

  • Could this topic’s impact extend across different Earth regions or populations?
  • Does it address universal needs or challenges that apply broadly?

Does it solve a major problem?

  • Does your topic address a significant barrier to space exploration or human survival?
  • Can it simultaneously solve pressing challenges on Earth, like resource scarcity or climate change?

Can it adapt?

  • Is your topic flexible enough to work in diverse environments on Earth and eventually on Mars?
  • Could it be modified or enhanced as technology evolves?

Will it inspire future work?

  • Does your topic create a foundation for further innovation?
  • Could it lead to spinoff technologies or applications?

The Feasibility Factor:

Turning Ideas Into Action

Is your topic realistic? Even the most transformative ideas need to be grounded in feasibility. This is about asking the practical questions. Great solution topics are ambitious but achievable within a defined scope.

  • Can measurable progress be made within 18 months?

  • Does it rely on existing tools and technology, or those likely available by 2027?

  • Is your topic specific, focused, and actionable?

  • Is it practical within budget, manpower, and material constraints?

  • Can it be scaled for use across regions or contexts?

  • Does it address a real-world problem with the potential for meaningful impact?

Potential markets

Field AI assistants have the potential to transform industries by providing real-time, autonomous support in challenging environments. By integrating AI into critical sectors such as manufacturing, disaster response, and remote exploration, we can improve efficiency, enhance safety, and drive innovation. Developing these systems now not only addresses current needs but also lays the groundwork for more resilient, adaptive technologies for the future.

Industrial IoT & smart manufacturing

  • Market Size: $439 billion global Industrial IoT (IIoT) market in 2024 (Precedence Research, 2024), projected to reach $847 billion by 2033 (IMARC Group, 2024).
  • Trends: Adoption of AI-driven edge computing devices for predictive maintenance, autonomous robotics, and real-time quality control in factories, mines, and oil rigs.
  • NASA Link: Stress-testing rugged AI in Earth’s harshest industrial environments (e.g., offshore drilling platforms) would mature autonomous systems for maintaining lunar habitat machinery and mining equipment.

Disaster response & emergency management

  • Market Size: $190 billion global disaster preparedness market (Precedence Research, 2024) and $167 billion emergency response sector (Mordor Intelligence, 2024)1 2 3.
  • Trends: AI tools for real-time damage assessment, resource allocation, and search-and-rescue coordination in earthquakes, wildfires, and floods.
  • NASA Link: Deploying portable AI in Earth’s disaster zones would refine algorithms for autonomous decision-making in lunar crises, like habitat breaches or solar storm warnings.

Remote exploration (polar, marine, desert)

  • Market Size: $14.2 billion combined global market (2024), comprising:
    • Polar Exploration: $9.8 billion | Polar Icebreaker: $8.2B1 and Polar Travel: $1.6B2 3
    • Marine Exploration: $3.3 billion | Marine Mining: $2.3B4 and Expedition Cruise Shipbuilding: $1.0B5
    • Desert Exploration: $1.1 billion | Aggregates for arid-region infrastructure: $591B total market6, with ~0.2% allocated to desert-specific projects7
  • Trends: AI systems for autonomous data collection and analysis in extreme environments, such as deep-sea research or Antarctic climate studies.
  • NASA Link: Validating AI in Earth’s remotest research stations would advance systems for lunar geology surveys and Mars analog missions, where real-time Earth collaboration is impossible.

Have your topic in mind?

Put your team together and let's pitch it.
register
REGISTER