Evolution of Autonomous Driving Technology

Autonomous driving technology is classified into various levels, with differing standards proposed by organizations like the U.S. Highway Traffic Safety Administration (NHTSA) and the Society of Automotive Engineers (SAE). While specifics in data division may vary, these classifications share commonalities in characterization. The transition from driver operation to autonomous vehicle driving begins at L3, marking a crucial juncture in autonomous driving technology's application.

Autonomous Driving Solutions:

Autonomous driving solutions branch into two routes: single-vehicle intelligence and vehicle-circuit coordination. Single-vehicle intelligence utilizes sensors like cameras, radar, and efficient algorithms to endow vehicles with autonomous driving capabilities. Conversely, vehicle-road synergy integrates people, vehicles, road information, and cloud-based systems, creating a "smart car + intelligent road" environment.

These routes aren't mutually exclusive; they complement each other. Vehicle intelligence forms the bedrock for realizing autonomous driving. Even in vehicle-road coordinated programs, vehicle intelligence remains indispensable. It serves as a redundancy system in cases where roadside intelligent facilities aren't accessible or fail, ensuring safe task completion. Additionally, vehicle intelligence acts as a critical component in vehicle-road cooperation, aiding system upgrades and new function development. This cooperation enhances data redundancy and safety, especially in complex traffic scenarios, amplifying sensing perspectives and bolstering autonomous driving's reliability and safety.

Considerations for Market Participants:

Apart from technical feasibility, market participants consider factors like participation rights, access thresholds, commercialization, and landing complexities in choosing their route. Currently, OEMs and autonomous driving solution providers predominantly opt for the single-vehicle intelligence route. This path offers a human-machine co-driving experience through functions like automatic parking and adaptive cruise control (L2+ functions), ensuring technology control while generating commercial profits. Tesla's FSD, Xiaopeng's NGP, and Azalea's NOA exemplify this approach adopted by leading manufacturers.

In scenarios like partially closed 2B and semi-closed settings, along with urban public service 2G environments, vehicle-road collaboration transforms extensive infrastructure to share perception and computation costs. This collaboration, driven by economic advantages and industrial development, catalyzes commercial opportunities.

Maturing Intelligent Driving Industry Chain:

The global intelligent driving industry chain comprises various industries in the upstream—sensors, chips, algorithms, high-precision maps—forming the foundation. Midstream activities see continuous development and R&D of intelligent driving products by OEMs through independent or collaborative initiatives. The service market arising from intelligent driving technology upgrades and operation holds a pivotal role, enhancing vehicles' autonomous driving capabilities. Unmanned delivery vehicles, networked car operations, and engineering vehicles streamline transportation, reducing costs and boosting efficiency for enterprises.