Recently, Carl Power CEO Wei Junqing announced, "Carl Power has taken the lead in achieving the world's largest-scale L4-level autonomous bulk commodity transport, surpassing 1 billion ton-kilometers, with a total demonstration fleet operation mileage exceeding 15 million kilometers."
At the same time, Carl Power has established its global innovation and R&D headquarters in the Kangbashi District of Ordos City, aiming to assist Ordos in becoming a benchmark city for integrated "vehicle-road-cloud" applications.
It is well known that China's automotive industry is moving towards the path of "vehicle-road-cloud integration." Academician Li Keqiang from the Chinese Academy of Engineering highlighted, "Vehicle-road-cloud integration is critical for promoting high-quality development in the intelligent and connected vehicle industry. It plays a key role in cultivating new productivity and shaping industrial momentum." Notably, Ordos City is one of China's first 20 pilot cities for vehicle-road-cloud integration, particularly excelling in coal mine logistics and future intelligent mining advancements.
Since 2021, Carl Power has conducted L4-level autonomous bulk transport operations in Ordos, representing a prime application of this technology in trunk logistics scenarios.
Under the framework of vehicle-road-cloud integration, Carl Power proposed the "Carl Navigation" autonomous driving hybrid-intelligence convoy technology based on vehicle-to-vehicle communication. This solution addresses the technical challenges that single-vehicle intelligence cannot overcome, enabling deployment and operation.
Currently, Carl Power has developed an overall architecture for vehicle-road and vehicle-cloud collaboration. It effectively integrates vehicle and road data through its self-developed KargoCloud intelligent logistics dispatch platform, which is compatible with local and national cloud platforms. This platform digitizes information about autonomous vehicle fleets, cargo, transport resources, and energy supplies. Leveraging robust data processing and intelligent dispatch capabilities, it achieves precise matching between transport demands and resources.
Through continuous testing in Ordos, Carl Power has built data and scenario barriers, accumulating extensive real-world scenarios and experienced truck driver data. This data is fed back into training and validating autonomous driving models, accelerating the evolution of autonomous driving technology through data-driven approaches and large-scale models.
Market validation has demonstrated the viability of Carl Power's hybrid-intelligence model. This model involves one human driver piloting an L2-level lead vehicle to guide multiple L4-level autonomous vehicles. This approach, where the lead vehicle navigates while trailing vehicles transition to full automation, improves safety by fivefold compared to human driving, reduces labor costs by 50–80%, and lowers overall operational costs by 20%.
The entire autonomous driving system costs between 100,000 and 150,000 RMB, and with unmanned operations and dual-shift vehicle schedules, it can save 200,000–300,000 RMB in labor costs annually. "Once you deploy hundreds of unmanned vehicles, investors and clients no longer question profitability," Wei Junqing revealed. Carl Power's self-operated autonomous fleet is already "profitable."
Carl Power is optimistic about the unmanned freight market. Wei Junqing stated, "We expect cross-regional L4-level autonomous driving to become a reality in the next 3–5 years, gradually rolling out nationwide." He believes, "The freight sector will be the first to achieve commercial breakthroughs in autonomous driving." Compared to last-mile delivery or robotaxi sectors, unmanned freight requires lower investment costs but has a higher route concentration, with profitability achievable by deploying 1,000–1,500 units.
As of December 20, Carl Power has a fleet of 300 autonomous vehicles and has taken the lead in achieving 1 billion ton-kilometers in L4-level autonomous bulk commodity transport-a key metric reflecting transport industry efficiency.
