Jeanette

onventional pitot-static air data systems remain a critical vulnerability for modern aircraft, particularly for uncrewed aerial systems (UAS), advanced air mobility platforms, and other small or highly autonomous vehicles. These legacy systems suffer from mechanical complexity, susceptibility to blockage and icing, single-point failure modes, and reliance on distributed processing architectures that increase size, weight, power, and integration complexity. This project proposes the development of a miniaturized, AI-enabled air data computer implemented as a system-on-chip (SoC) that replaces traditional pitot-static hardware and external microprocessors with a compact, tightly integrated sensing and estimation platform.

The proposed SoC architecture integrates distributed flush-mounted MEMS pressure sensors, on-chip analog-to-digital conversion, signal conditioning, digital signal processing, and physics-informed machine learning algorithms within a single computational framework. By embedding air data estimation, fault detection, and adaptive calibration directly within the SoC, the system minimizes reliance on external microcontrollers while enabling low-SWaP, fault-tolerant, and scalable air data computation. Spatial pressure gradients measured across the aircraft surface are fused using hybrid physics-based and data-driven estimation techniques to produce real-time estimates of airspeed, acceleration, altitude, vertical speed, angle of attack, angle of side-slip, accompanied by confidence and health metrics.

Phase I will focus on system-level modeling and partial hardware-in-the-loop validation of the SoC-based architecture, including algorithm development, pressure sensor interfacing, and embedded estimation pipelines. Laboratory-scale demonstrations will establish feasibility, quantify robustness improvements relative to conventional architectures, and validate the benefits of SoC integration in reducing system complexity and external processing requirements. The resulting technology provides a clear pathway to a fully integrated flight-ready SoC in Phase II, with strong commercialization potential across civil, defense, and emerging aviation markets while advancing national priorities in aviation safety, autonomy, and intelligent sensing.

Biography:

Jeanette Bryan Hariharan is a Principal AI and Computer Vision Engineer at Space-Tech with extensive experience applying artificial intelligence and machine learning to solve real-world engineering challenges. Her work focuses on developing advanced systems for pattern recognition, object localization, anomaly detection, and industrial optimization, combining signal and image processing with scalable AI deployment. With a background spanning industry and academia, she has held engineering and leadership roles at organizations including Lockheed Martin, The Timken Company, Meggitt, Safran, Rockwell Automation, and Aerosonic, as well as faculty positions at Florida Gulf Coast University and Ave Maria University. Her expertise bridges AI innovation, control systems, embedded software, and robotics applications, delivering solutions that enhance automation, safety, and operational efficiency.

Jeanette holds a Ph.D. in Electrical Engineering and is a Certified Professional Engineer. She is passionate about advancing intelligent technologies that drive productivity and meaningful real-world impact

Address:United States