Chongqing Launches L3 Highway Autonomous Driving Rules

Chongqing, one of China's largest municipalities, has officially released new regulations governing Level 3 and above autonomous vehicle testing on public highways, marking a significant step in the country's push to commercialize advanced self-driving technology. The rules, jointly issued by 3 municipal agencies on May 7, establish a full-process management framework designed to balance innovation with public safety.

The regulation, titled the 'Chongqing Intelligent Connected Vehicle Highway Testing Management Rules (Trial),' was published jointly by the city's Economic and Information Commission, Public Security Bureau, and Transportation Commission. It aims to fill a critical regulatory gap for high-speed autonomous driving scenarios, setting the stage for broader deployment of L3+ self-driving systems across China's highway networks.

Key Takeaways From the New Regulation

• Scope: Covers Level 3 (conditional automation) and above autonomous vehicles on designated highway segments
• Joint oversight: 3 municipal agencies — economic affairs, public security, and transportation — share regulatory responsibility
• Full lifecycle management: Rules govern every phase from application and testing to incident response and data reporting
• Industry alignment: Designed to support Chongqing's broader intelligent connected new energy vehicle (ICNEV) strategy
• Commercialization focus: Explicitly aimed at accelerating the transition from testing to real-world commercial deployment
• Trial basis: Published as a 'trial' regulation, signaling room for iteration as the technology evolves

Why Highway Testing Matters for Autonomous Driving

Highway testing represents a fundamentally different challenge compared to urban autonomous driving. Vehicles travel at speeds exceeding 60 mph, leaving far less time for decision-making. Lane changes, merging traffic, and high-speed obstacle avoidance demand faster sensor processing and more robust AI models.

Until now, most Chinese autonomous vehicle testing has focused on urban roads and controlled environments. Chongqing's new framework directly addresses what regulators call a 'system gap' — the absence of formal rules for testing advanced autonomous systems in high-speed scenarios. This gap has slowed companies from validating their L3+ systems under realistic highway conditions.

The distinction between Level 2 and Level 3 autonomy is critical here. At Level 3, the vehicle — not the driver — is responsible for monitoring the driving environment. This shift in liability and technical responsibility demands entirely new regulatory frameworks, which is exactly what Chongqing is now providing.

Chongqing's Strategic Position in China's AV Race

Chongqing is no newcomer to autonomous vehicle development. The city has long positioned itself as a hub for China's intelligent connected vehicle ecosystem, leveraging its mountainous terrain and complex road networks as natural proving grounds for self-driving technology.

The municipality is home to major automotive manufacturers including Changan Automobile, one of China's largest state-owned automakers, which has been aggressively developing L3 and L4 autonomous systems. Other players in Chongqing's AV ecosystem include partnerships with tech giants like Baidu, Huawei, and numerous startups.

Several factors make Chongqing uniquely suited for highway AV testing:

• Complex topography: Mountain roads, tunnels, and bridges create diverse testing conditions rarely found elsewhere
• Dense highway network: The city operates one of southwestern China's most extensive expressway systems
• Government support: Local authorities have consistently prioritized ICNEV development as a pillar industry
• Existing infrastructure: The city already operates smart road infrastructure with V2X (vehicle-to-everything) capabilities on select corridors

How This Compares to Global AV Regulations

Chongqing's move mirrors a broader global trend of governments racing to establish regulatory frameworks for advanced autonomous driving. In the United States, the National Highway Traffic Safety Administration (NHTSA) has taken a largely hands-off federal approach, leaving much regulation to individual states like California, Arizona, and Texas.

California's Department of Motor Vehicles (DMV) currently permits autonomous vehicle testing on public roads, including highways, but requires detailed reporting of disengagements and accidents. By contrast, Chongqing's approach appears more centralized, with 3 agencies sharing oversight in a coordinated framework.

In Europe, the UN Economic Commission for Europe (UNECE) approved regulations for L3 automated lane-keeping systems (ALKS) in 2021, but implementation has been slow. Germany became the first country to allow L3 vehicles on public roads when Mercedes-Benz received approval for its DRIVE PILOT system in late 2021, though it remains limited to speeds under 40 mph on specific highway segments.

Chongqing's willingness to open high-speed highway testing for L3+ vehicles suggests China may be moving faster than Western counterparts in creating real-world testing environments. This aggressive posture could give Chinese automakers a significant data advantage in training and validating their autonomous systems.

The Technical Challenges of L3+ Highway Driving

Achieving reliable Level 3 autonomy on highways requires solving several interconnected technical challenges that push current AI systems to their limits.

Perception at speed is perhaps the most fundamental issue. LiDAR, cameras, and radar sensors must identify and classify objects at distances exceeding 200 meters, with processing latencies measured in milliseconds. At highway speeds, a vehicle covers roughly 30 meters per second, meaning any delay in perception can have catastrophic consequences.

Decision-making under uncertainty also becomes more complex. Highway scenarios involve predicting the behavior of multiple vehicles simultaneously — will the truck ahead brake suddenly? Is the car in the adjacent lane about to merge? These predictions rely on sophisticated AI models trained on massive datasets of real-world driving behavior.

Key technical requirements for L3+ highway autonomy include:

• Redundant sensor arrays: Multiple overlapping LiDAR, camera, and radar systems to ensure no single point of failure
• High-definition mapping: Centimeter-accurate maps updated in near real-time for lane-level positioning
• V2X communication: Vehicle-to-infrastructure connectivity for receiving traffic data, construction alerts, and emergency warnings
• Fail-safe protocols: Automated minimum risk condition (MRC) maneuvers that safely stop the vehicle if the system encounters an unresolvable situation
• Over-the-air updates: Continuous software improvement based on fleet-wide learning

What This Means for the Global AV Industry

Chongqing's new regulations send a clear signal to automakers, technology companies, and investors worldwide. China is systematically removing regulatory barriers to advanced autonomous driving testing, creating a more permissive environment for companies to collect the highway driving data essential for improving their systems.

For Western companies like Waymo, Cruise, and Tesla, this development intensifies competitive pressure. Chinese firms operating under these new rules will accumulate highway testing miles and edge-case data that directly improve their AI models. Tesla's Full Self-Driving (FSD) system, which recently expanded its supervised highway capabilities, faces growing competition from Chinese rivals who may soon have access to more diverse high-speed testing environments.

The commercial implications are also significant. L3 autonomy on highways is widely seen as the first viable mass-market application of advanced self-driving technology, because highways are more structured and predictable than urban streets. Companies that can validate and commercialize L3 highway systems first will capture enormous market value — analysts at McKinsey have estimated the global autonomous driving market could reach $400 billion by 2035.

Looking Ahead: From Testing to Deployment

Chongqing's trial regulations represent a beginning, not an endpoint. The 'trial' designation indicates that regulators expect to refine the rules based on real-world testing outcomes, safety data, and industry feedback.

Several developments are likely in the coming months. First, expect major Chinese automakers — particularly Changan and potentially BYD and NIO — to announce highway testing programs under the new framework. Second, the data generated from these tests will likely inform national-level regulations that could standardize L3+ highway testing across all of China.

The broader trajectory is unmistakable. China is building a comprehensive, layered regulatory infrastructure for autonomous driving — from urban robotaxis to highway passenger vehicles. As Chongqing joins cities like Beijing, Shanghai, and Shenzhen in establishing advanced AV testing rules, the country is positioning itself as the world's largest laboratory for self-driving technology.

For global industry observers, the message is clear: the race to commercialize Level 3+ autonomous driving is accelerating, and the regulatory landscape is becoming as important as the technology itself.