At the AutoCC 2025 Intelligent Driving Navigation Ecological Innovation Summit held on August 28th, several intelligent driving practitioners held a roundtable discussion on the topic of "Future Imagination Space of Large Models and Navigation Intelligent Agents".
During the discussion, Yang Yuekun, a senior navigation expert at NIO, and several other guests pointed out that in the era of big models, intelligent driving has become increasingly dependent on the navigation ecosystem. Yang Yuekun believes that with the increasing level of intelligence of intelligent agents such as ADAS and cockpit, there are more data and collaborative parts that navigation needs to provide, and their importance increases accordingly.
Huang Haifeng, the person in charge of Zhiji Auto's intelligent driving map, pointed out that promoting end-to-end and AI big models is the trend of the industry and cannot be avoided. For example, when encountering problems such as road changes, weight restrictions, and unexpected situations, using end-to-end and large-scale models to output directly from perception performs better. At the same time, he also emphasized that navigation maps still play a key role in AI intelligent driving.
TomTom solution architect APAC lead Liu Tianjun pointed out that in the era of big models, navigation needs to face challenges such as timeliness, accuracy, safety, and reliability of computation. In this regard, industry partners such as OEMs and suppliers need to work together to promote the next stage of assisted driving.
From the perspective of high-precision positioning, the person in charge of Qianxun Location Intelligent Driving stated that regardless of the stage, the absolute positioning requirements of intelligent driving for vehicles only differ in importance and tasks undertaken. Qianxun position is adapting to industry trends and continuously optimizing and adjusting related products.
