Recently, the State Administration for Market Regulation and the Standardization Administration jointly published China’s first national standards for grading automated driving. The standards will take effect in March next year and will be a benchmark for automobile manufacturers to develop their technologies.
The Taxonomy of Driving Automation for Vehicles provides official definitions for self-driving cars from zero (L0) to five (L5). At the lower end of the scale (L1-L2), drivers still need to largely control the vehicle, but as the vehicle moves up on the certification scale (L3-L5), it requires less input from the driver. L5 is considered fully automated. Intuitiveness and attention to detail convey the automation level of vehicles accurately and raise public awareness on automation level.
The Society of Automotive Engineers’ (SAE) definition is a bit ambiguious as it defines L2 as “partial automated driving” and L4 level as “high-level automated driving”. On the contrary, the Chinese version of classification is more detailed and clear.
China’s L0, L1 and L2 levels require that the driver and the automated driving system work together on detecting and responding to objects and events while the SAE version only requires drivers to perform the tasks for these levels.
L3 is named as Conditionally Automated Driving as it could continuously perform all dynamic driving tasks under its designed operating conditions.
L4 is called highly automated driving. At this level, the vehicle is also able to takes measures to reduce the risk of accident should the driving automation system fail to perform the necessary actions.
Finally, L5 is fully automated driving. This level means that there are no limitations by any operating design scope, and the vehicle can continuously perform all dynamic driving tasks under any conditions. Should the driving automation system fail to perform dynamic driving tasks, the L5 system takes measures to reduce the risk of accident and to minimize any damages.
In addition to the Taxonomy of Driving Automation for Vehicles, the Ministry of Industry and Information Technology promulgated the “Automobile Incident Data Recording System” a few months ago, and plans to require M1 vehicles to install the automobile incident data recording system starting on January 1, 2022.
The recent policies might be a precursor to the government further strengthening the supervision of smart cars and automated driving. A much more strict, clear and meticulous automated driving environment might boost the technology’s commercial viability and mass deployment of automated vehicles in China.