As everyone is liking what’s happening in Space with regards to Brainchip I’ve just posted quite a few new things going on there over in the NASA page with this one being the best
www.sbir.gov
Edge AI Platform for Space and Unmanned Aerial Imagery Intelligence
Award Information
Agencyepartment of Defense
Branch:Air Force
Contract:FA8649-23-P-0196
Agency Tracking Number:FX224-OCSO1-1566
Amount:$74,941.00
Phasehase I
Program:SBIR
Solicitation Topic Code:X224-OCSO1
Solicitation Number:X22.4
Timeline
Solicitation Year:2022
Award Year:2023
Award Start Date (Proposal Award Date):2022-11-01
Award End Date (Contract End Date):2023-02-04
Small Business Information
Misram LLC
1 Paddington Lane
Holmdel, NJ 07733-1111
United States
DUNS:078446162
HUBZone Owned:No
Woman Owned:No
Socially and Economically Disadvantaged:Yes
Principal Investigator
Name: Vidya Sagar
Phone: (201) 912-9568
Email:
vidya@spectronn.com
Business Contact
Name: Rajarathnam Chandramouli
Phone: (847) 345-4731
Email:
mouli@spectronn.com
Research Institution
N/A
Abstract
Global persistent awareness is critical for the Air Force. Warfare is becoming increasingly complex due to congested and contested operational environments, maneuvering targets, and rapidly changing threats. The Air Force seeks on-demand awareness of adversary actions anywhere on the globe by securely gathering, processing, and fusing multiple types of trusted data, including satellite and aerial imagery. The Joint and Allied Force networks will share processed intelligence to rapidly identify threats and targeted intelligence. Executing AI algorithms for satellite and aerial imagery fusion solely on terrestrial cloud infrastructure will incur severe latency due to limited data rate between a satellite and ground station, wireless link packet losses, delays due to revisit times, etc. Therefore, we propose to extend our commercial distributed edge AI sensor fusion software stack to satisfy the needs of Air Force customers. A 1000X latency reduction and 5X improvement in intelligence inference accuracy will be demonstrated.