embedUR Systems has engineered a compact Ultra-Wideband (UWB) sensing solution for vehicles, built on NXP Semiconductors’ Trimension NCJ29D6 platform. The innovation introduces gesture-based control for cars, enabling functions such as hands-free trunk opening and smart parking assistance using real-time motion detection.
The embedded solution operates with an AI model sized at just 215 KB, delivering fast, efficient gesture recognition directly on NXP’s low-power UWB transceiver. Designed for automotive environments, the NCJ29D6 chip also supports features like Digital Key access, Child Presence Detection, and kick sensors—all integrated into a single, low-energy platform.
embedUR Systems has tailored its gesture and motion recognition engine specifically for this platform, optimizing it for both performance and energy efficiency. The compact design and edge-based processing make the solution ideal for real-world automotive use, eliminating the need for larger, power-hungry processors.
This development comes at a time when the automotive industry is seeing rising interest in contactless interactions and intelligent in-vehicle sensing. The technology also offers potential applications in consumer electronics, industrial automation, and smart home ecosystems, opening doors beyond the transportation sector.
Eric Smiley, Vice President of Business Development at embedUR Systems, stated, “Our focus on embedded AI and UWB has allowed us to create user-friendly, real-time solutions that improve driver convenience and safety. This project highlights the potential of UWB technology when paired with lightweight AI for everyday use.”
Michael Leitner, General Manager of Secure Car Access at NXP Semiconductors, added, “embedUR’s skill in wireless sensing and efficient edge processing has allowed them to unlock powerful capabilities from our UWB platform. Their work on gesture detection has shown promising results, and we see strong potential for future applications.”
With a track record of over 20 years, embedUR Systems has consistently delivered high-performance software for connected devices and edge platforms. The company specializes in developing AI models that operate within strict memory and power budgets—routinely optimizing models between 200 and 500 KB without sacrificing inference speed or accuracy.
In addition to this automotive breakthrough, embedUR’s portfolio includes solutions like real-time image segmentation on Arm Cortex-M7 processors and low-footprint facial recognition on sub-1MB memory devices. The company also provides tools like ModelNova, a development resource for rapid Edge AI prototyping aimed at helping product teams accelerate intelligent system design.
embedUR Systems, headquartered in Silicon Valley, is a premier provider of embedded solutions, AI, and edge computing technologies. With over 20 years of experience, the company has been instrumental in accelerating product development for top-tier telecom providers, network equipment manufacturers, and silicon vendors. Leveraging deep expertise in AI/ML, IoT, and advanced networking, embedUR Systems powers millions of connected devices globally, driving the evolution of next-generation intelligent systems.
