First of all, we need to clarify the classification of autonomous driving levels. According to the standards formulated by the Society of Automotive Engineers (SAE), autonomous driving is divided into 6 levels from L0 to L5. L0 is manual driving, without any assisted driving functions; L1 is auxiliary driving, which can provide single-function assistance such as adaptive cruise control (ACC) or lane keeping assist (LKA); L2 is partial autonomous driving, which can realize the combination of multiple functions such as adaptive cruise control and lane centering control, and the driver still needs to pay attention to the road conditions at all times; L3 is conditional autonomous driving, which can realize autonomous driving in specific scenarios (such as highways), and the driver can temporarily leave the steering wheel, but needs to take over the vehicle in time when the system prompts; L4 is high-level autonomous driving, which can realize autonomous driving in most scenarios, and the driver does not need to take over the vehicle in normal conditions; L5 is full autonomous driving, which can realize autonomous driving in all scenarios, regardless of weather, road conditions and other factors, and the vehicle can complete the driving task without any human intervention. At present, the autonomous driving technology of most mainstream automobile manufacturers is in the L2 to L2+ stage. Models such as Tesla Model 3/Y, BYD Han, and BMW 5 Series are equipped with advanced L2+ level autonomous driving functions, which can realize automatic following, lane changing, automatic parking and other functions, greatly reducing the driving burden of drivers. Data show that by the end of 2025, the penetration rate of L2-level autonomous driving models in the global new car market has reached 45%, and the penetration rate in China has exceeded 50%, becoming the mainstream configuration of mid-to-high-end models. However, it is still a long way from L2 to L5. The key bottlenecks restricting the development of high-level autonomous driving technology mainly include three aspects: first, sensor technology. Autonomous driving vehicles need to rely on sensors such as cameras, lidar, radar and ultrasonic sensors to perceive the surrounding environment. At present, the cost of lidar is still high, and the perception accuracy of sensors in complex weather (such as heavy rain, heavy fog, snow) and complex road conditions (such as construction sections, rural roads) still needs to be improved; second, algorithm technology. The core of autonomous driving is the algorithm, which needs to process a large amount of sensor data in real time, make accurate judgment and decision-making. At present, the algorithm still has deficiencies in dealing with unexpected situations (such as sudden crossing of pedestrians, sudden braking of vehicles in front), and the ability of “human-like thinking” needs to be further improved; third, laws and regulations and ethical issues. At present, most countries have not formulated perfect laws and regulations for high-level autonomous driving, such as the definition of liability in the event of an accident, the privacy protection of driving data, etc. At the same time, ethical issues such as “who to protect when facing a dilemma” also need to be solved. In terms of policy support, various countries are actively promoting the development of autonomous driving technology. The United States has formulated a clear autonomous driving development plan, allowing L3-level autonomous driving vehicles to be put on the road in some states; the European Union has issued relevant regulations to clarify the legal status of autonomous driving vehicles; China has listed autonomous driving as a key development direction of the new energy automobile industry, and has carried out pilot projects in cities such as Beijing, Shanghai and Guangzhou, accumulating a lot of practical experience. Industry insiders predict that L3-level autonomous driving vehicles will be widely put on the road around 2028, and will be mainly used in scenarios such as highways and urban expressways; L4-level autonomous driving vehicles will be popularized in specific fields such as logistics and public transportation around 2035; and L5-level full unmanned driving may not be fully popularized until around 2040. Of course, this time node will be affected by factors such as technological progress, policy support and market acceptance. For consumers, the popularization of autonomous driving technology will greatly change the way of travel, making driving safer, more efficient and more comfortable. For the automobile industry, autonomous driving technology will promote the transformation and upgrading of the industry, and the competition pattern of the industry will also change significantly. In the future, with the joint efforts of governments, enterprises and research institutions, the dream of full unmanned driving will eventually come true.
