Research Design for the Wet-gap Crossing Problem
Abstract
Project Title: Research Design for the Wet-gap Crossing Problem Project Abstract The research design of this project will use the machine learning and computer vision capabilities of the Jetson Nano and the Isaac SDK installed on the NVIDIA Kaya robot to enhance the control and manipulation of the environment performed under human (player) guidance. The goal is to complete the wet-gap crossing within limited time. The final research design will be refined through experiments for establishing the best performing design. The project has four main phases: hardware construction and software setup, student training, software development; and data collection, testing and experimentation. The specific details of some phases are subject to change based on what the team discovers through experimentation. The software programs developed for each type of robot Ð the boat, ground vehicle and tank will be semi-autonomous and will comply with the rules of the competition to maximize the possibility of the successful wet-gap crossing within the time limit. This project is in the area of team-based collaborative robotics. Collaborative robotics promises to solve a wide variety of problems that are of importance to society and the armed forces. Collaborative robotics can also allow robots to improve their abilities, by learning through trial and error and then conveying this information to the others in the team; allowing all involved to become smarter at a faster rate.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 24, 2020
- Source ID
- W911NF2010044
Entities
People
- Sambit Bhattacharya
Organizations
- Army Contracting Command
- Fayetteville State University
- Office of the Secretary of Defense