The chip was developed in collaboration with GlobalFoundries, marking POLYN’s transition to more advanced semiconductor nodes for current and future product generations. The development signals a broader industry shift toward embedding intelligence directly into sensing hardware to enable faster and more efficient data processing at the edge.
POLYN’s VibroSense solution processes high-frequency vibration data captured from tires to estimate the peak friction coefficient in real time. By applying neural network models trained on diverse driving conditions, the system is designed to provide early insight into road grip, enabling predictive responses before events such as braking or loss of traction occur.
According to the company, the technology has been validated through controlled braking tests and benchmark comparisons, demonstrating alignment with real-world friction behavior across a range of surfaces, including asphalt, concrete, and wet road conditions. The system is engineered to maintain consistent performance in complex environments, such as uneven terrain or aquaplaning scenarios.
The chip operates in conjunction with a tire pressure monitoring system (TPMS) system-on-chip, using accelerometer-based vibration signals to assess tire-road interaction. By embedding sensing capabilities directly within the tire, the approach seeks to improve accuracy over indirect estimation methods commonly used in current vehicle systems.
At the core of the technology is POLYN’s neuromorphic analog signal processing architecture, which utilizes asynchronous analog neurons to process data with low latency and minimal power consumption. This enables deployment in space- and energy-constrained automotive environments, where real-time responsiveness is critical.
The company indicated that it is engaging with Tier 1 suppliers and automotive original equipment manufacturers to explore integration of the technology into future vehicle platforms.
