On May 8, 2026, under the cooperation bridge built by the China-CEEC Innovation Cooperation Research Center (hereinafter referred to as “the Research Center”), the School of Mechanical and Automotive Engineering of Ningbo University of Technology successfully held an online technology exchange meeting with universities and research institutions including the Hungarian National Research Network (HUN-REN) and E?tv?s Loránd University. The two sides conducted in-depth exchanges on intelligent tire sensing, road perception, intelligent driving and other fields, laying a solid foundation for follow-up scientific research cooperation and achievement transformation.

During the exchange, the Hungarian team focused on three core technological achievements: the intelligent tire sensor system, road surface anomaly detection, and road surface classification and recognition. They shared key technologies including tire force perception based on 3D piezoresistive sensors, tire-road friction coefficient estimation, and lightweight AI model deployment. Drawing on their respective research strengths, both sides held in-depth discussions on sensor packaging and integration, signal denoising, real-vehicle testing and verification, and industrial application scenarios. Both sides agreed that the relevant technologies boast broad application prospects in intelligent driving, active safety, intelligent transportation, and other fields, and share a sound cooperation foundation with Ningbo University of Technology in vehicle engineering, intelligent detection equipment, and related disciplines.

The Research Center has given full play to its role as an international scientific and technological cooperation platform and a bridge. In the early stage, it carried out precise matchmaking around the technological advantages and cooperation needs of both parties, actively coordinated resources between Chinese and Hungarian universities and research institutions, and facilitated the efficient implementation of the exchange event. Going forward, the Research Center will continue to follow up on the cooperation intentions of both parties, further advance technological exchanges toward joint R&D, talent cultivation, and achievement transformation, and help deliver practical and tangible cooperation outcomes.

Promotion of Hungarian Technological Achievements for Cooperation

Project 1: AI-powered Smart Tire Sensor System

This system embeds piezoresistive force sensors inside tires and combines them with lightweight neural networks. It can recognize road type and quality in real time, as well as estimate tire forces and friction coefficients. The technology has reached the TRL5 prototype stage and has been validated on roads in Budapest and multiple vehicle models. It features high precision, low computing power consumption, and strong robustness, which can improve driving safety and autonomous driving efficiency. Top-tier journal papers have been published and patents have been filed for this technology.

Project 2: Intelligent Tire Measurement System Equipped with 3D Piezoresistive Force Sensors

It is used for road abnormality detection. The sensor features high sensitivity and can collect three-dimensional force signals of tires. Features are extracted through adaptive Hermite function transformation, and classification is performed by combining the analytical threshold method and the VP-NET lightweight neural network. The system has been verified on roads in Budapest through wireless transmission and real-vehicle tests, with an average accuracy of over 97% for road abnormality detection. This solution features low computational power demand and strong robustness, making it suitable for vehicle-mounted embedded platforms. It can be applied to pothole detection, road condition assessment, and intelligent driving control. Patents have been granted and achievements have been published in IEEE journals.

Project 3: Road Surface Classification Technology Based on Piezoresistive Force Sensors and Time-Frequency Analysis

By embedding 3D force sensors into tires to collect deformation signals, features are extracted using a variety of time–frequency distributions (MWSP, WV, ZAM, etc.), and classification is carried out via thresholding, MLP, VGG16, and VP-NET. In real-vehicle tests on the Nissan Leaf, the model achieves high accuracy in detecting abnormal road surfaces, and the detection results are highly correlated with road quality. The solution is lightweight and robust, adapting to variable environments without retraining. It can be used for driving scene perception and adaptive speed control in autonomous vehicles. The research has been published in the IEEE Access journal.

If you are interested in the above projects, please contact:

cell phone：+86-17280096277

E-mail: cooperation@cceec.tech