The 2025 Spring Festival Intelligent Driving Crowd-Testing Report: NOA System Still Needs Improvement in Urban Areas, While Some Models Show Outstanding High-Speed Stability

The 2025 Spring Festival Intelligent Driving Crowd-Testing Report: NOA System Still Needs Improvement in Urban Areas, While Some Models Show Outstanding High-Speed StabilityDuring the 2025 Spring Festival, Zhihu, in collaboration with the Intelligent Connected Vehicle Branch of the China-Europe Association, released the "2025 Spring Festival Intelligent Driving Crowd-Testing Report." Based on data from Zhihu's "2025 Spring Festival Intelligent Driving Crowd-Testing Competition," the report comprehensively evaluated the intelligent driving systems, particularly the NOA (Navigation on Autopilot) systems, of 14 mainstream intelligent vehicles

The 2025 Spring Festival Intelligent Driving Crowd-Testing Report: NOA System Still Needs Improvement in Urban Areas, While Some Models Show Outstanding High-Speed Stability

During the 2025 Spring Festival, Zhihu, in collaboration with the Intelligent Connected Vehicle Branch of the China-Europe Association, released the "2025 Spring Festival Intelligent Driving Crowd-Testing Report." Based on data from Zhihu's "2025 Spring Festival Intelligent Driving Crowd-Testing Competition," the report comprehensively evaluated the intelligent driving systems, particularly the NOA (Navigation on Autopilot) systems, of 14 mainstream intelligent vehicles.

The crowd-testing involved over 10,000 kilometers of testing equivalent to crossing China twice spanning more than 150 hours. NOA accounted for 86.9% of the total mileage. Tests covered over 30 diverse scenarios, including urban and highway driving, day and night conditions. Data was recorded by CHEK intelligent driving equipment, resulting in this detailed report.

The report reveals an overall satisfaction rate of 82.35% for the participating intelligent driving systems. However, only 17.65% of participants expressed complete trust in urban NOA functionality. This highlights the challenges of current intelligent driving technology in complex urban environments. On average, the dangerous takeover mileage for all NOA systems was 220.21 kilometers, meaning a driver takeover was required approximately every 220 kilometers.

Significantly, these dangerous takeovers were not evenly distributed across road conditions. While urban NOA mileage only constituted 25.96% of the total NOA test mileage, a considerable 77.14% of dangerous takeovers occurred in urban NOA scenarios. This indicates that the stability of current NOA systems in complex urban traffic environments is relatively low and requires further optimization to enhance safety.

The report detailed performance differences across various vehicle models and scenarios. In urban driving, models like the Zhiji LS6, Wei Brand Blue Mountain, and Tengshi Z9GT demonstrated higher stability. The Zhiji LS6 exhibited the best performance in terms of NOA dangerous takeover mileage, reaching 149.49 kilometers. On highways, the Tengshi Z9GT, Xiaomi SU7 Pro, and Lynk & Co 09 stood out, with the Tengshi Z9GT achieving a remarkable 351.10 kilometers before a dangerous takeover.

While most vehicles could operate in complex urban areas, the report noted that system responsiveness and processing capabilities still require optimization in challenging environments. Zhihu user "Zz's Intelligent Life," after participating in the crowd-testing, pointed out that current intelligent driving assistance systems significantly reduce driver burden on highways, making long-distance driving feasible. However, end-to-end models introduce uncertainties, necessitating continued driver supervision.

Another Zhihu user, "Wu Wang," observed that while intelligent driving technology has made impressive progress, its development hasn't met widespread expectations. He emphasized that current intelligent driving systems remain in the "assisted" driving phase and cannot fully replace human drivers.

Hao Jingshan, deputy director of the Shenzhen Autonomous Driving Research Center, stated that the development of intelligent driving technology relies on real consumer feedback. The vast data from this crowd-testing will help manufacturers optimize their systems, ultimately transitioning intelligent driving from assistance to complete automation. Gao Ling, secretary-general of the Data Application and Consumer Research Branch of the China Automobile Dealers Association, highlighted that evaluating intelligent driving performance using real-user data enhances objectivity, transparency, and consumer acceptance of intelligent vehicles.

Professor Zhu Xichan from the Automotive Engineering College of Tongji University emphasized the significance of this crowd-testing. He argued that the mass production of high-level intelligent driving systems with NOA features has changed the methods of evaluating automotive functions. Traditional single-scenario testing in test fields cannot comprehensively assess the generalization ability of NOA systems in the ever-changing environment of real-world road traffic. The "2025 Spring Festival Intelligent Driving Crowd-Testing Competition" provides real-world user feedback, a testing methodology worthy of promotion.

In conclusion, the "2025 Spring Festival Intelligent Driving Crowd-Testing Report," based on over 10,000 kilometers of real-world road tests, provides an objective assessment of intelligent driving system performance. The results highlight progress in intelligent driving technology while also pointing out shortcomings, particularly in complex urban environments. This is crucial for the continuous development and improvement of intelligent driving technology and provides consumers with valuable information for selecting intelligent vehicles. The report's detailed analysis of performance differences across models and scenarios offers consumers more refined choices and valuable improvement directions for automakers. Future progress, driven by technological advancements and accumulated consumer feedback, will hopefully improve the safety and stability of intelligent driving systems, ultimately realizing the vision of fully autonomous driving. However, consumers must maintain a clear understanding of the real capabilities of intelligent driving systems and exercise necessary caution and supervision. The success of this crowd-testing provides valuable experience for future evaluations of similar intelligent driving systems, signaling a new trend toward more scientific, objective, and transparent assessment models. This will not only accelerate the development of intelligent driving technology but also enhance consumer trust and acceptance.

Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])