In the Daliang Mountains of Sichuan and a county in Inner Mongolia, they spotted a 3-trillion-yuan unmanned vehicle business opportunity | Exclusive from 36Kr

Text | Li Anqi

Editor | Li Qin

Early in the morning, on a rural road in Liangshan, Sichuan, a box-shaped unmanned vehicle was cruising at a speed of 40 km/h at the foot of the mountain. The unmanned vehicle set off from an express delivery outlet in the county town. After traveling for about an hour and covering nearly 30 kilometers, it successfully arrived at the township express delivery station.

After completing two trips of transportation every day, before returning to the county in the afternoon, the unmanned vehicle can also transport the freshly picked fruits of local fruit farmers in need back to the county and then send them to all parts of the country.

Not only in Liangshan, Sichuan, but also in pastoral and agricultural areas in Inner Mongolia, Longnan in Gansu, Pingxiang in Jiangxi and other regions, Jiushi's unmanned vehicles are quietly changing the daily lives of local people.

This is not only a story about how technology changes lives, improves express delivery efficiency and optimizes the experience, but also a trillion-yuan business.

Kong Qi, the founder and CEO of Jiushi, told 36Kr that there will be a 10% replacement rate in the annual market of 30 million commercial vehicles, that is, 3 million vehicles. Calculated at an average price of 70,000 yuan per unmanned vehicle, the market scale will reach 200 billion yuan. Coupled with problems such as the aging workforce and the labor gap in the freight industry, the market for unmanned urban distribution can reach trillions.

The primary investment market has smelled the prospects. 36Kr Auto exclusively learned that the autonomous driving company "Jiushi Intelligence" recently completed the delivery of a $100 million Series B3 financing. So far, the total cumulative financing in the Series B has been nearly $300 million.

The Series B financing of Jiushi Intelligence was jointly led by CDH Baifu, Blue Lake Capital, a leading US dollar fund and a large listed company, with follow-on investments from Asia Investment Capital, Wuxi Venture Capital, Yongxin Ark, a logistics industry player, BV Baidu Ventures, C&D Emerging Investment and Unicorn. This is a rare single-round financing of over $100 million in the primary market of autonomous driving in recent years.

It is understood that the funds raised in this round by Jiushi Intelligence will be used for the continuous R & D of the next generation of unmanned vehicles, the establishment of its own supply chain system, the expansion of domestic and international markets, the construction of urban operation networks, and the establishment of an industrial ecological alliance.

CEO Kong Qi told 36Kr Auto that so far, Jiushi has cumulatively delivered over 3,000 unmanned vehicles, landed in about 200 domestic cities and expanded to 6 countries. By the end of this year, the delivery scale of Jiushi's unmanned vehicles will reach 10,000 units.

"Unmanned vehicles need to address pain points for customers to pay the bill."

Founded in 2019, Jiushi Intelligence is a technology company focusing on the R & D and application of L4-level autonomous urban distribution products. Its founder and CEO, Kong Qi, was an early member of Baidu's unmanned vehicle team. After graduating with a master's degree from the Department of Computer Science at Shanghai Jiao Tong University in 2011, Kong Qi joined Baidu's Shanghai R & D Center. At the end of 2013, he participated in setting up Baidu's unmanned vehicle team at its US R & D Center.

Kong Qi recalled to 36Kr Auto that at that time, he was mainly responsible for the decision-making, planning and control modules of autonomous driving, and jointly "crafted" Baidu's first batch of unmanned vehicles with the early team.

In 2018, Kong Qi, who was already a member of Baidu's Technical Committee, joined JD.com and built a unmanned delivery team from scratch. In 2021, Kong Qi decided to start his own business and founded Jiushi Intelligence.

Since 2013, the autonomous driving industry has experienced several investment booms and industry bubbles. On the one hand, many technical experts from Silicon Valley in the United States have returned to China to start businesses and joined the wave of unmanned driving, and the industry has attracted a lot of capital. On the other hand, the industrialization of autonomous driving has been tortuous, with founders suing each other in court and a number of autonomous driving companies going out of business.

Low-speed unmanned driving is a scenario that is easier to implement in the eyes of engineers, but it is also a track that is considered to have a relatively low technical threshold by geniuses and experts. However, in Kong Qi's view, a few years ago, autonomous driving was in the stage of algorithm verification and evolution. If it wanted to be circulated in the market as a commodity, it still lacked the ability to turn technology into products.

He believes that autonomous driving in the logistics field can more easily continuously obtain data and evolve under commercial conditions.

Another group of peers share the same idea: start - ups such as Neolix, White Rhino and Haomo.AI. Large companies such as JD Logistics, Cainiao Logistics and Meituan have also entered the market.

However, the implementation of low - speed unmanned vehicles is not as simple as expected. Kong Qi told 36Kr Auto that its unmanned vehicles not only need to drive on complex urban and rural roads, but also need to be implemented in scenarios such as snow - covered areas in the Northeast, deserts in the Northwest, mountain roads in Yunnan and pastoral areas in Inner Mongolia, posing quite diverse technical challenges.

At the same time, obtaining the "right - of - way permit" for unmanned vehicles to drive on public roads requires breakthroughs in each city, and the high cost of vehicle hardware can only be reduced when the supply chain gradually matures.

Kong Qi told 36Kr Auto that Jiushi provided a similar service in the early days, using unmanned vehicles to transport goods for clients. "The income from one trip was 40 - 50 yuan." Currently, some peers are acting as operators, providing autonomous driving capacity and services. The technologies of large companies such as JD.com and Cainiao are mostly piloted in their own logistics scenarios.

However, under this model, the Jiushi team found that the model of operating its own fleet still requires a lot of human resources to support the technology.

Therefore, Jiushi took another path.

In 2023, after Jiushi launched its first - generation Z5 series of vehicles, the person - in - charge of an express delivery outlet in Shaanxi contacted the Jiushi team through business registration information and insisted on buying the whole vehicle. "We didn't even have a formal purchase contract at that time. The two sides discussed and hand - wrote a contract together." Kong Qi recalled when talking about Jiushi's first customer to buy the whole vehicle.

From the real feedback of subsequent customers, Jiushi found that if it could really address the pain points of customers in goods delivery, they would be more willing to buy the whole vehicle.

Kong Qi told 36Kr Auto that according to the original business model, the cost of hiring a van and a driver for an express delivery outlet was about more than 10,000 yuan per month. Calculated over 5 years, the software and hardware cost of an unmanned vehicle is about 3,000 yuan per month, and a Z5 can carry 600 - 800 express parcels per trip.

For example, in Liangshan, Sichuan, the distance between counties and townships often exceeds 30 kilometers. Coupled with the shortage of the working - age population, the efficiency of logistics was not high under the previous transportation model of manpower and vehicles. Sometimes, express parcels had to be accumulated for 2 - 3 days before being delivered together, and the timeliness of local villagers' agricultural product shipments was also low.

At the beginning of this year, the person - in - charge of a local express delivery outlet purchased 4 Jiushi Z5 unmanned vehicles. The maximum cruising range of the vehicles can reach 180 kilometers, and they can make multiple round - trips between the township and county stations. "Now the unmanned vehicles can make 2 trips a day, and the efficiency has improved a lot."

According to the preliminary calculation of this outlet, through Jiushi's unmanned vehicles, the operating cost of the outlet has decreased by more than 40%. The income increase of local fruit farmers is also relatively obvious. Similarly, in a county in Inner Mongolia, the delivery of 2 Jiushi Z5 unmanned vehicles can reduce the cost of an express delivery outlet by 56.25% per month.

Taking the unmanned vehicle business to the townships, "Don't underestimate customers' determination to make money."

Unmanned vehicles can reduce customers' operating costs, but new technologies sometimes also mean high thresholds.

Jiushi also found in the research that the unit price of low - speed unmanned vehicles used to be often more than 200,000 yuan per vehicle. "Some customers have tight cash flow, and it's actually a bit difficult for them to pay more than 200,000 yuan to buy a car at once. So we launched a subscription - based payment model."

In June 2024, when Jiushi launched multiple models such as the Z2, Z5 and Z8, it provided a software payment plan. For example, the selling price of the best - selling Z5 model is 49,800 yuan, but customers can pay the unmanned driving software fee quarterly at 7,000 yuan per quarter.

This has greatly reduced the threshold for customers to buy cars, and Jiushi's customers have begun to be distributed in various regions across the country. For example, in the Northeast region, express delivery outlets use unmanned vehicles to transport express parcels to township stations. This not only protects couriers from the cold winter wind, but also ensures the timeliness of consumers' online shopping. In Wuxi, Jiangsu, Jiushi's unmanned vehicles also set off at 4 am to transport fresh fruits and vegetables from wholesalers to canteens in urban and township units.

With the trial and promotion of the first batch of users at the end of 2023, Jiushi began to have repeat customers, and some small and medium - sized customers also actively approached after seeing the printed advertisements on the unmanned vehicles.

New business models are also emerging, and some customers are gradually developing into the role of operators. Kong Qi told 36Kr that some customers may only use the vehicles 3 - 4 times a day, and they will rent out the vehicles during their spare time. "Now some customers buy unmanned vehicles in batches and operate an autonomous - driving version of Huolala. These customers have become unmanned driving operators."

"Customers have more potential than we thought and know better how to operate unmanned vehicles in the scenario." Kong Qi said.

These are all the "explosion" signals of the large - scale implementation of low - speed unmanned vehicles that Kong Qi has seen: there is a spontaneous demand for buying cars in the market, and there is a secondary demand fission. Unmanned vehicles can truly solve customers' problems in delivery and freight, rather than just being pilot demonstration cases in industrial parks.

Jiushi said that currently, it has developed about 100 distributors and nearly 300 service providers across the country. It is worth mentioning that by the end of 2024, 45 of the distributors were developed from Jiushi's unmanned product customers, and 25 distributors are in the process of transforming into urban operators.

On the other hand, as the number of real customers increases, the speed of obtaining "road - going permits" for Jiushi's unmanned vehicles is also getting faster.

Kong Qi told 36Kr that currently, Jiushi has obtained road - going qualifications in 29 provinces and over 100 cities across the country, and has implemented the regular operation of unmanned vehicles with partners. In nearly one - third of these cities, the road - going permit documents or licenses were obtained with the help of customers who pushed local government departments to complete the procedures.

"The most difficult first step in getting franchisees on board is to make them recognize your product. After you can help them make money, things actually become simpler. Don't underestimate customers' determination to make money."

The technology of low - speed unmanned vehicles also aims to catch up with Tesla.

Of course, the breakthroughs in the market and policy also depend on the gradual maturity of low - speed unmanned driving technology.

In terms of vehicle design, Jiushi's unmanned vehicles have no cockpits and no safety drivers. From the real - time operation pictures shown by Kong Qi to 36Kr Auto, the vehicles can drive on public roads, semi - closed roads and industrial park roads. More than 3,000 vehicles can accumulate more than 100,000 kilometers of data every day. The scenario library covers diverse situations such as snow - covered areas in the Northeast, deserts in the Northwest, mountain roads in the Southwest and pastoral roads in Inner Mongolia.

Kong Qi said that this is due to the fact that Jiushi not only delivers to large - scale enterprise brands, but also has a large number of small and medium - sized customer groups. "The transportation, terrain and weather in the geographical areas where these groups are located all have their own characteristics, which also pose greater challenges to the generalization ability of unmanned vehicle products and technologies.

Jiushi said that its unmanned vehicles use 4 types of perception sensors, including lidar, cameras, millimeter - wave radars and ultrasonic sensors. After strict environmental and reliability tests, the unmanned vehicles can work stably for more than 5 years in the temperature range of - 30°C to + 55°C. In terms of software, based on the OCC (Occupancy Network) fusion detection technology, multi - scenario pre - decision - making algorithm and dynamic distributed scheduling framework, Jiushi can compress the inference delay to less than 50 milliseconds, greatly improving the adaptability to complex urban roads.

In addition, Jiushi also plans to launch a light - map L4 autonomous driving capability in the second half of this year.

Currently, Jiushi's unmanned vehicles still need to use high - precision maps and need to be pre - deployed before being delivered to customers for use.

Kong Qi told 36Kr Auto that after upgrading to the light - map solution, high - precision maps will basically not be used on some structured roads. Only in some special intersections or scenarios with relatively blurred boundaries will some high - precision map elements be used. "It is expected that the pre - deployment time can be compressed to 3 - 4 days."

It is worth noting that since Tesla launched the wave of intelligent driving assistance in the passenger car field, the supply chain of intelligent vehicles has also been maturing rapidly. Kong Qi told 36Kr Auto that the new - generation L4 - level urban distribution vehicles launched by Jiushi this year are equipped with automotive - grade components such as solid - state lidar. With the continuous large - scale implementation of the products, the cost of the whole vehicle can be further reduced in the future.

In Kong Qi's view, under the multiple stimuli of the market, policy, technology and supply chain, the current unmanned urban distribution market is in an explosive stage.

Currently, Jiushi Intelligence has 4 models, namely the Z2, Z5, Z8 and Z10. As of last year, Jiushi had accumulated more than 10,000 on - hand orders, cumulatively delivered more than 3,000 vehicles and served more than 600 customers. In the first quarter of 2025, Jiushi said that its new order volume increased by more than 4 times year - on - year.

In terms of customer types, Jiushi's customers are not limited to small and medium - sized customers. There are also large - brand customers in different scenarios such as China Post, STO Express, Sinopharm Medical Devices, Giti Tire and Papa Tang, covering fields such as express delivery, fresh food supermarkets and pharmaceutical cold chain.

For example, Jiushi has completed the delivery of the first batch of unmanned vehicle orders to Giti Tire, and will gradually launch the unmanned tire delivery service in 32 cities in the future. Through unmanned vehicles, Jiushi can help Giti Tire complete the replenishment and transfer tasks from urban warehouses to auto repair shops, and can also meet the special needs for emergency tire use at night.

In the overseas market, Jiushi's unmanned vehicles are also exploring markets in Singapore, Japan, South Korea, the Middle East, etc. It has obtained the first unmanned logistics vehicle license in Singapore, and its unmanned vehicles have also entered the supply chain systems of local leading enterprises such as FairPrice in Singapore.