Intelligent Doppler Radar-based Tracking and Classification of Target Objects
Abstract
Overall Goal This project bridges Minority Serving Institute (MSI) and DoD, specifically initiating several research and educational collaborations between Virginia Commonwealth University (VCU) and Naval Surface Warfare Center (NSWC) Dahlgren Division. Although these two campuses are located within 70 miles, not so many collaborations have been achieved yet. The new project and activities described in this proposal aim to establish research and educational collaborations between VCU Engineering College and NSWC. The role of partners at NSWC is to advise and evaluate the project s progress so that the PI can get direct feedback from real-world problems at DoD/ONR. The project details and directions will be adaptably revised based on the interests and needs of the sponsor, DoD/ONR. The PI supports the existing STEM student employment program at NSWC Dahlgren [1] that enhances the educational experience, and provides financial aid to help the student s educational goals. The PI himself will participate in ONR Summer Faculty Research Program (SFRP) [2]. Objectives Both MSI undergraduate and graduate students will be recruited from underrepresented minorities of various backgrounds (such as gender, age, and race) to actively conduct the research project on a specific basic science relevant to long-term national security needs [15]. They are working together to reach the project outcomes and gather educational experience through the real-world project described in the proposal. The PI will integrate the research and education through the MSI program to increase the number of graduates in STEM to support the success of minority students. Research Abstract The problems of existing radars are inaccuracy and unreliability for early detection and warning of incoming threats and a high accuracy estimate of their position and dynamics. The proposed solution is to adopt state-of-the-art artificial intelligence technologies with the latest Doppler radar. The Doppler effect produces velocity data about objects at a distance by bouncing a microwave signal off a desired target and analyzing how the object s motion has altered the frequency of the returned signal in the form of a spectrogram. A specific feature range for Doppler radar data will identify anomaly events and target objects acquired in real scenarios as shown in Fig. 1. With the latest machine learning approaches, the spatiotemporal variation gives direct and highly accurate measurements of the radial component of a target s velocity relative to the radar. The project will deliver a proof of concept on an intelligent range Doppler with software implementation on training, testing, and self-learning. The main three threats are 1) Designing digital signal processing of spectrogram used for Spatial-temporal clustering, 2) Implementing deep learning algorithms for target detection and tracking by modifying relevant machine learning toolboxes, and 3) Evaluating the performance beyond the initial simulation setting into potential defense applications. Fig.1. Expected visualization outcomes of radar-grams overlay the outdoor scene associated with anomaly monitoring a wider range of distance
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 28, 2023
- Source ID
- W911NF2310161
Entities
People
- Yuichi Motai
Organizations
- Army Contracting Command
- Office of the Secretary of Defense
- Virginia Commonwealth University