Research Rovers: AI Research Assistants for NASA

Help NASA assess emerging capabilities for AI-based research assistants. #science

$30,000 in prizes
oct 2023
293 joined


Like many of us who depend on technical and scientific literature in our own work, NASA researchers need to understand the state of research in a particular domain or multiple domains. They are also often faced with the daunting task of becoming familiar with a body of literature that they don't have prior experience with. Whether you are an academic doing your own research, a data scientist exploring new machine learning techniques, or anyone else delving into a field of study that's new to them, this problem may feel very familiar. With the Research Rovers challenge, NASA sought innovative approaches to help assess emerging capabilities in AI-based research assistants.

The Solution

In this challenge, participants developed and demonstrated an AI-based research assistant solution to be used by NASA researchers with data sources like arXiv, Google Scholar, PubMed, OpenAlex, CrossRef, and the NASA Technical Reports Server (NTRS).

For their submissions, they provided a 5-10 minute video demonstration and short writeup describing their solution and its potential for assisting NASA researchers in their work. Making a submission was a fairly significant undertaking that drew on a variety of skill sets. We are grateful to everyone who participated!

By the end of the challenge, 28 teams had made final submissions, which were then scored by a 10-person judging panel composed of researchers and digital transformation leaders at NASA. Judges evaluated submissions based on their relevance, effectiveness, deployability and novelty.

The Results

Participants explored a wide range of approaches to this problem, many of them leveraging recent advances in generative AI using large language models (LLMs). Some of the most impressive solutions combined multiple research assistant features including literature summarization, chatbots for user-specified research papers, author searches and more into a single application.

See the results announcement for more information on the winning approaches and the teams who developed them.