SBIR-STTR Award

Enhancing MQ-9 Aircrew Situational Awareness by Automated Image Recognition from Aerial Images, based on Explainable-AI
Award last edited on: 3/1/2021

Sponsored Program
SBIR
Awarding Agency
DOD : AF
Total Award Amount
$800,000
Award Phase
2
Solicitation Topic Code
AF192-001
Principal Investigator
Saied Tadayon

Company Information

Z Advanced Computing Inc

11204 Albermyrtle Road
Potomac, MD 20854
   (301) 294-0434
   info@zadvancedcomputing.com
   www.zadvancedcomputing.com
Location: Single
Congr. District: 06
County: Montgomery

Phase I

Contract Number: FA8649-19-P-A343
Start Date: 8/2/2019    Completed: 8/2/2020
Phase I year
2019
Phase I Amount
$50,000
ZAC's detailed image recognition technology based on our Explainable-AI (artificial intelligence) is capable of obtaining various fine details from images taken at various camera angles, including aerial images from MQ-9 (as well as satellite or other aerial images), under varying conditions and occlusion, to accurately locate and identify objects of interest and enhance aircrew situational awareness and performance. Our technology has surpassed the best alternatives in the industry (ResNet).

Phase II

Contract Number: FA8629-20-C-5021
Start Date: 11/13/2019    Completed: 11/13/2021
Phase II year
2020
Phase II Amount
$750,000
In this SBIR Phase II project, ZAC detailed image recognition solution, based on Explainable-AI (artificial intelligence), will be adapted to the USAF needs: It will provide detailed intelligence and full picture from satellite or aerial/drone images to warfighters, to enhance drone aircrew situational awareness and performance. The solution will automatically and accurately locate, recognize, and identify objects (e.g., tanks, trucks, people), targets and potential threats, with details, in real-time, from any view direction, from aerial images (e.g., from MQ-9) under varying conditions. For example, it will automatically and accurately identify the detected airplane as C-130, and it will provide decision quality information to end-users, such as MQ-9 aircrew, ISR community, and tactical forces, to minimize risks and better execute and achieve their missions. In addition, ZAC's Explainable-AI image recognition has the advantage of using less computational power, while being modular in nature, making it capable of being deployed in the field or at the edge (i.e., on-board the UAV), as it will occupy a smaller footprint/weight, and require less battery power due to lower power consumption.