SBIR / STTR #24.4

Army SBIR solicitation for precision eye tracking sensors and synthetic data solutions for AI/ML

This Army SBIR opportunity seeks innovative solutions from small businesses for two distinct technology areas: - Precision contact lens eye tracking sensors for Extended Reality (XR) applications - The Army Applied SBIR Office is requesting feasibility studies and prototype development for embedded contact lens eye-tracking systems - Prototypes must collect binocular eye-gaze data, pupil size, and blinks, and be compared to high-precision trackers - Broader applications may include health monitoring and smartphone replacement, leveraging semiconductors, micro-LEDs, sensors, and RF/Non-RF data transfer - Notable requirements: small business eligibility, adherence to SBIR guidelines, and rugged, mobile form factor suitable for Army environments - Synthetic data generation and domain adaptation for AI/ML in computer vision - Surveyed products include ShapeNet 3D Model Database, SYNTHIA Synthetic Dataset, and SceneNet RGB-D Dataset - Services requested: synthetic data generation, domain adaptation using GANs, and privacy-preserving synthetic data for sensitive applications - Requirements include high-fidelity labeling, multi-modality support, domain transfer, and realism in simulation environments - No specific part numbers or purchase quantities are provided for the eye tracking sensors; datasets are referenced by name - Potential bidders for the eye tracking sensor project include Epitome Research Innovations Inc., Intellisense Systems, Inc., Merge Plot LLC, Nielson Scientific LLC, and Zansors LLC - Funding for Phase I projects is limited to $250,000 for a 6-month performance period - Place of performance and delivery is primarily at Army and Department of Defense facilities in Arlington, VA

Description

This solicitation aims to advance methods for generating and labeling synthetic data representing various classes of Radio Frequency (RF) signals to support training of Electronic Support and Signals Intelligence (SIGINT) models. The goal is to enhance automated detection, characterization, and identification of Signals of Interest (SoI) using AI/ML techniques. The project addresses challenges in managing the increasing volume and diversity of RF signals to reduce operator workload and improve battlespace awareness and decision-making. The effort includes phases focusing on direct-to-Phase II proposals, synthetic data generation for training detection models, and demonstrating AI/ML efficacy with potential dual-use applications.

View original listing