Syracuse, New York – BosonQ Psi Federation LLC (BQP) of Syracuse, New York, received a SpaceWERX Small Business Innovation Research (SBIR) award for its development of quantum-inspired artificial intelligence applications designed to improve space domain awareness. The company did not disclose the prize amount.
This project will focus on the development and validation of physically constrained quantum-assisted machine learning (PC-QAML) software applications aimed at improving the identification and classification of unknown objects and behaviors in the Earth’s orbit environment.
Space Domain Awareness organizations monitor thousands of objects in orbit, including operational satellites, debris, and other objects that may require additional analysis. BosonQ Psi Federation says its technology is designed to address the challenges associated with uncorrelated tracks, which are detections that cannot be immediately matched to known objects.
Related: Researchers focus on next-generation transistor technology for quantum, photonic, bio/organic integrated circuits
Physics meets quantum
The company says its PC-QAML approach combines physically-based modeling and quantum-inspired computational techniques to reduce the computational requirements of artificial intelligence models while maintaining accuracy. This technology is designed for deployment on edge computing platforms where processing power, power consumption, and communication availability may be limited.
According to BosonQ Psi Federation, its approach reduces the size of AI models from approximately 14 million parameters to approximately 2,000 parameters while maintaining a classification accuracy of over 99%. The company also reports up to 10x reduction in inference latency and approximately 90% reduction in power consumption compared to traditional machine learning approaches.
The technology was previously demonstrated on the NVIDIA Jetson Nano edge computing platform in the Space Domain Awareness TAP Lab and evaluated for orbital separation detection at the SDA Mini Accelerator in 2025, according to the company. BosonQ Psi Federation has identified PC-QAML as a candidate technology for future uncorrelated track classification and threat simulation catalog integration.
BosonQ Psi Federation develops software solutions based on physics-based computing and quantum-inspired machine learning techniques for aerospace, defense, and national security applications. The company’s BQPhy platform is designed to support engineering and scientific applications by combining physically-based models with advanced computing approaches.
BPQ says the SpaceWERX award represents the company’s entry into the federal market and will work with government officials to support further development and validation of PC-QAML applications.
