Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Session: 1
Room: Private Dining
Carley Brinkley - Fayetteville State University
Co-Author(s): Antwaun Tune, Fayetteville State University, Fayetteville, NC; Eric Griep, Fayetteville State University, Fayetteville, NC.
With the increasing efforts to advance scientific understanding of Mars, research needs to be conducted to collect data. With human travel to Mars being very challenging, the National Aeronautics and Space Administration (NASA) has sent robotic exploration devices to collect data such as pictures and surface crust samples. Bringing data and results allows NASA to test hypotheses and develop scientific theories about Mars. Our project is to simulate the Martian surface by using the gaming engine, Unity. Robots can be simulated in this environment with the Robot Operating System (ROS) and Unity bridge to imitate the process of collecting data from Mars and reporting it back to NASA. Simulation in robotics is an integral part of the design process, most notably in space robotics where you often only get one chance for everything to work properly. Photorealistic simulation for AI is important for collecting data used in the training of computer vision models. This gives AI the ability to complete tasks like object detection and vision-based navigation. Unity’s game engine allows for the seamless usage of realistic 3D objects from sources like Quixel Megascans. Our simulation can create a crater similar to the ones found on Mars; the environment includes sand-like meshes, photorealistic rock walls, and rock patches. The collaboration between our team and other teams within the research program is feeding into a larger project. We’re working forward to establishing a business that finds and authenticates data using trained AI robotics. The trained AI can support jobs we can and would do on Earth and outer space. Some tasks can be too dangerous, time-consuming, and impossible for humans; however, AI robotics can carry out these tasks and possibly deliver more accurate results. By participating in this research project, we’ll gain more knowledge about the exploration of extraterrestrial bodies like Mars. We’ll also be able to help NASA with other projects like the deep space exploration Artemis mission.
Funder Acknowledgement(s): This work was supported in part by a NASA MUREP Space Technology Artemis Research (M-STAR) Implementation Grant.
Faculty Advisor: Dr. Samit Bhattacharya, sbhattac@uncfsu.edu
Role: I worked with the co-authors in creating the environment. My portion of this research dealt with the terrain. I had to retrieve images and videos from official NASA rovers that are on Mars. The purpose was that the material provided the team with realistic ideas of what to build. We met weekly, and I documented our progress and setbacks. With the information from these meetings, I was able to write the abstract with the help of the co-authors.