U.S. Strategic Operations Command is seeking technology providers capable of delivering “cutting-edge computer vision capabilities” for detection and classification for all USSOCOM unmanned systems.

The Problem

According to USSOCOM, computer vision “rarely performs as intended,” and results in missed detections or improperly classified objects. Challenges also exist with obtaining training data, utilizing the models on constrained resources, and deploying models over the air.

The Solution

USSOCOM is seeking solutions for a “computer vision inference engine” and model training solution that can autonomously detect, classify, and adapt to new targets and environments. This would enhance the operational effectiveness in resource-constrained and communication-denied environments.

The computer vision inference engine and model retraining solution would need to be:

• Robust — Needs to demonstrate robust object detection and classification capabilities.

• Optimized — For size, weight, power, and cost for deployment on a variety of platforms.

• Scalable — Should enable training and adaptation to new objects and environments.

• Modular — Should offer a modular architecture suitable for “seamless integration” with existing architectures.

• Autonomous — Needs to enhance the autonomous capabilities of unmanned systems operating in challenging environments.

Awards

Financial details were not released, but USSOCOM is seeking solutions it can downselect into negotiated awards.

Submissions

Israeli startups with relevant capabilities can submit their capability for review using this form. Deadline is Oct. 13. There’s a white paper template available that you can use, as well a list of the criteria that USSOCOM will use to evaluation solutions.

Next Steps

USSOCOM is hosting a virtual Q&A teleconference on Oct. 3, which is the day after Yom Kippur. You can click here to sign up for the Q&A.

Difficulty

Simple white paper submission.

Questions

Contact Scott Cohen at The CET Sandbox at scott@cetsandbox.com.