Zhuoyu Technology Releases Pansing Intelligent Assisted Driving Solution for Commercial Vehicles

From "Chengxing" to "Panxing," Zhuoyu Technology Brings the Same Mobile Intelligence Base into Commercial Vehicle Scenarios

Zhong Lilei, 1 hour ago. Zhuoyu's commercial vehicle business has entered a new stage of mass production.

On the evening of October 29, Zhuoyu Technology held a commercial vehicle technology launch event in Zhuhai themed "New Journey," officially releasing its commercial vehicle intelligent assisted driving solution — Zhuoyu Panxing™ — and systematically disclosing the latest progress in its commercial heavy trucks, buses, light trucks, ROBOVAN and global business.

As a new business vertical that Zhuoyu is fully pushing toward mass production following its passenger vehicle business, the commercial vehicle business has achieved a closed loop from technology to commercialization. The release of Panxing marks the further expansion of Zhuoyu's "Intelligence for All Mobility" strategy from its passenger vehicle base into commercial vehicle scenarios, driving the large-scale implementation of mobile physical AI across multiple verticals. From "Mobile Intelligence Base" to "Intelligence for All Mobility," Zhuoyu's commercial vehicle business enters a new stage of mass production

Over the past few years, the intelligent driving industry has moved from the rule-based era into the data-driven era, and is accelerating toward the foundation model era. At the end of 2025, Zhuoyu was the first to propose the concept of a "Mobile Intelligence Base"; in April 2026, Zhuoyu released its "Intelligence for All Mobility" strategy, driving mobile physical AI from "one intelligence per vehicle" toward "one intelligence shared by all things," and advancing the layout and implementation of multi-vertical businesses including passenger vehicles, commercial vehicles and unmanned logistics vehicles.

Zhuoyu Technology CEO Shen Shaojie said at the launch event that the commercial vehicle business is a new business vertical that Zhuoyu is fully pushing toward mass production following its passenger vehicle business, and has also achieved a successful closed loop from technology to commercialization; it is important proof of Zhuoyu's practice of its mission to "provide the world with safe and relaxed mobile intelligence."

Over the ten years since its founding, Zhuoyu has developed full-stack in-house capabilities covering a complete hardware suite, model algorithms, a data-driven engine, engineering implementation capability, and safety and process infrastructure, and possesses the capability for large-scale, million-level system delivery. These capabilities have become Zhuoyu's confidence for navigating cycles and advancing "Intelligence for All Mobility."

Passenger vehicles are the base; "Chengxing" capabilities migrate to commercial vehicles to form "Panxing"

Zhuoyu's mobile intelligence capabilities first came from long-term validation in the passenger vehicle field. At the launch event, Zhuoyu disclosed the latest progress of its passenger vehicle business: Zhuoyu currently has 22 passenger vehicle customers, with cumulative mass-produced models exceeding 60, designated models exceeding 110, and has recently secured cooperation with automaker customers including Toyota and Hyundai.

On the technology side, the high-comprehension large model of Zhuoyu's passenger vehicle intelligent assisted driving business brand "Chengxing" has undergone four major version evolutions; version 4.1 has already been deployed on multiple models, and the upcoming 4.2 version will be launched in the fourth quarter of this year. Zhuoyu's native multimodal foundation model is planned for mass production deployment in the first quarter of 2027. In the passenger vehicle L3 business, Zhuoyu released its L3 architecture solution at the Beijing Auto Show in April this year, and is simultaneously advancing L3 pilot licenses in multiple cities, with major progress expected in the first half of 2027.

Following the passenger vehicle solution "Chengxing," Zhuoyu named its commercial vehicle solution "Panxing." "Pan" means the solidity of rock; "Xing" means steady and far-reaching progress. Shen Shaojie said that "Chengxing" represents ease and efficiency, ready to set off at any time, while "Panxing" represents a steadfast commitment to safety, arriving wherever it goes. From Chengxing to Panxing, Zhuoyu has successfully migrated Chengxing's mobile physical AI capabilities to commercial vehicles, letting the same "brain" grow in two different "bodies."

Panxing includes two versions, PDS 4 and PDS 5, covering trunk logistics and urban intelligent mobility

At the launch event, Zhuoyu officially disclosed the version system of the Panxing solution.

Among them, Zhuoyu Panxing™ PDS 4 corresponds to the high-comprehension large model version 4.0 and above, supports highway NOA capability, and primarily meets the needs of commercial heavy truck trunk logistics scenarios; it has already achieved mass production in multiple brands and reached a mature delivery state.

Zhuoyu Panxing™ PDS 5 corresponds to Zhuoyu's native multimodal foundation model; in addition to highway NOA, it will also support full-scenario capabilities such as urban NOA, targeting urban intelligent mobility for commercial buses, light trucks and other forms, and offering greater room for future upgrades.

Cooperation reached with the TOP6 Chinese commercial heavy truck brands; leading customers validate mass production capability

At the launch event, Zhuoyu announced that Panxing has reached cooperation with the TOP6 Chinese commercial heavy truck brands, including: XCMG Automotive, Shaanxi Automobile Group, Sinotruk, FAW Jiefang, Foton Motor and Dongfeng Commercial Vehicle. Together, the six brands account for more than 93% of the domestic commercial heavy truck market.

Among them, XCMG is both a Zhuoyu customer and a Zhuoyu shareholder. XCMG was the first to raise with Zhuoyu the need for "can commercial heavy trucks also use good intelligent driving," which became one of the important triggers for launching the Panxing business. On September 20, 2026, the XCMG Hanjing new energy heavy truck was officially released, and the mass-produced model equipped with Zhuoyu Panxing™ PDS 4 is expected to be delivered by the end of this year.

Shaanxi Automobile is Panxing's first customer to achieve mass production deployment and operation. In June 2026, the first DeLong X5000 equipped with Zhuoyu Panxing™ PDS 4 officially entered mass production, and hundreds of vehicles are currently running on Yunda Express trunk lines. In the past two months, the usage rate of intelligent assisted driving has risen rapidly, and it is expected that by December this year, more than 90% of highway mileage will use highway NOA.

Sinotruk and Zhuoyu have reached all-domain cooperation covering both domestic and overseas markets, and have cooperated on designating multiple models; it is expected that by the end of this year, Sinotruk's high-end heavy truck SITRAK model equipped with Zhuoyu Panxing™ PDS 4 will roll off the mass production line.

FAW Jiefang has designated multiple models including the J7, J6 Pro, Yingtu, and Yinghang, and the Jiefang J7 Chuangling Edition equipped with Zhuoyu Panxing™ PDS 4 was first shown to the public at this year's Beijing Auto Show. Zhuoyu also joined hands with FAW Jiefang at the IAA Transportation in Hanover, Germany, to showcase their co-developed models and solutions to the global commercial vehicle industry. Among them, the J6 Pro is one of the first mass-production models equipped with Zhuoyu Panxing™ PDS 4.

Foton Motor and Zhuoyu have reached a strategic cooperation on all-domain intelligent driving for commercial vehicles. The two parties will jointly develop intelligent driving solutions adapted to scenarios such as commercial heavy trucks, light trucks, and unmanned logistics vehicles, and jointly expand overseas markets including the Middle East, South America, and Australia-New Zealand. The first Auman Galaxy 5 equipped with Zhuoyu Panxing™ PDS 4 will be launched to the market by the end of this year.

On the Dongfeng Commercial Vehicle side, the Dongfeng Tianlong KH and KL models have been cooperatively designated. In early October this year, the Dongfeng Tianlong KH L2+ intelligent tractor was released, equipped with Zhuoyu Panxing™ PDS 4, and is expected to enter mass production by the end of this year. From commercial heavy trucks to buses, light trucks, and VANs, Panxing achieves multi-category coverage of commercial vehicles.

In addition to commercial heavy trucks, Panxing is also being deployed in multiple scenarios such as buses, light trucks, and VAN platform models.

In the bus sector, Zhuoyu has reached in-depth cooperation with Yutong Bus, and is simultaneously working with Guoqi Zhikong to expand across multiple models, jointly promoting the application and deployment of high-level intelligent driving technology in the bus sector. The Yuwei E9 PRO City Edition will be equipped with Zhuoyu Panxing™ PDS 5, based on the NVIDIA Thor platform and applying Zhuoyu's native multimodal foundation model, supporting not only highway NOA but also urban NOA.

In the light truck and VAN sectors, the all-new light truck and VAN platform models developed by Zhuoyu in cooperation with BAIC Foton will also be equipped with Zhuoyu Panxing™ PDS 5, supporting urban NOA. From trunk-line logistics to urban distribution, from heavy trucks to light trucks, Zhuoyu is writing "intelligence for everything that moves" into more commercial vehicle scenarios.

In terms of globalization, Zhuoyu has reached overseas cooperation with partners including Sinotruk, Foton Motor, and Yutong Bus, with products covering all categories of models such as commercial heavy trucks, buses, and light trucks. In the future, Zhuoyu will work with customers to bring commercial vehicle intelligent assisted driving to the world.

At the launch event, Shen Shaojie summarized the progress of the Panxing commercial vehicle business as the "four alls": full coverage of product types, full adaptation to power types, full cooperation with leading brands, and full layout both domestically and overseas. He said that Zhuoyu can confidently say: "For commercial vehicle assisted driving, choose Zhuoyu Panxing."

Zhuoyu chose to take the lead in driving mass production deployment of commercial vehicles not only out of business judgment, but also from a comprehensive understanding of user value, economic and social value, and commercial value in the commercial vehicle industry.

At the launch event, Shen Shaojie mentioned how the team was moved by the book "Rolling Wheels" written by a reporting team from Sanlian Life Weekly. The book records the real stories of nearly 20 truck drivers: those hauling vegetables in Shouguang, those hauling coal in Shanxi, those running the Qinghai-Tibet route, and drivers engaged in China-Russia cross-border transport. These stories made the Zhuoyu team realize that in the mind of every truck driver, a precise "algorithm" is deployed: planning routes, scheduling rest, calculating fuel costs, and even every press of the accelerator is precisely calculated.

But in reality, commercial vehicle drivers have long faced long driving hours, high-intensity attention, complex road conditions, and physical wear and tear. In China, there are more than 30 million commercial vehicle drivers with similar experiences. Therefore, Zhuoyu believes there is "no reason not to create" a safe, relaxed, and money-saving intelligent assisted driving system.

From the perspective of economic and social value, commercial vehicles are the main artery of China's economy. Over the past year, China's freight volume reached 58.7 billion tons, of which nearly three-quarters was completed by commercial vehicle drivers driving trucks on highways. Logistics costs are highly correlated with enterprise costs and consumer costs, and improving the safety and efficiency of trunk-line transportation has important social value.

From the perspective of commercial value, China's commercial heavy truck market has entered an era of stock competition and structural replacement of old vehicles with new ones, and intelligent upgrading has become a natural opportunity for differentiation. At the same time, in 2025 the scale of commercial vehicle exports exceeded one million, and China's commercial vehicle going global has formed a full-industry-chain pattern of "complete vehicles + components + services," which also provides important opportunities for Zhuoyu and its partners.

Unified foundation, scenario enhancement, and engineering adaptation solve the five major challenges of intelligent assisted driving for commercial heavy trucks

Achieving "safe, relaxed, and money-saving" is not easy. Liu Sikang, Vice President of Commercial Vehicle Business at Zhuoyu Technology, said that intelligent assisted driving for commercial heavy trucks is more difficult than for passenger cars, reflected in at least five aspects: little model data, high perception requirements, low safety fault tolerance, great difficulty in vehicle control, and extreme cost sensitivity.

A fully loaded heavy truck weighs about 40 tons and travels at 90 km/h on highways, with a long body, many blind spots, and a long braking distance; the center of gravity changes greatly between empty and full loads, and combined with high-load operation, all of these bring real technical challenges.

Facing these problems, Zhuoyu did not build a separate system specifically for heavy trucks, but adopted a three-sentence methodology: unified foundation, scenario enhancement, and engineering adaptation.

First, unified foundation, that is, migrating the foundation model from the passenger car platform without reinventing the wheel. Passenger cars and commercial vehicles both mainly operate on paved roads, and their underlying capabilities such as road understanding, identification of traffic participants, and risk judgment share commonalities.

Second, scenario enhancement, that is, feeding in dedicated heavy truck data so that the model better understands heavy trucks and enhances its understanding of trunk-line transportation scenarios.

Finally, engineering adaptation, that is, ensuring large-scale mass production through systematic engineering capabilities such as perception, dynamics, safety redundancy, coordination of braking and retarders, and HMI.

Liu Sikang said: "We enable the same 'brain' to learn to drive commercial heavy trucks."

The industry's first mass-produced commercial heavy truck end-to-end large model, supporting the robust deployment of highway NOA

At the model level, Zhuoyu successfully migrated its passenger vehicle intelligent driving model to heavy truck scenarios, inheriting over 3 billion kilometers of accumulated passenger vehicle data and nationwide mass-production generalization capability, and enhanced it with 400 million frames of heavy-truck-specific data. Liu Sikang said that Zhuoyu has broken the notion that "end-to-end can only be deployed in passenger vehicles" and successfully achieved the industry's first mass-produced commercial heavy truck end-to-end large model.

This model enables commercial heavy truck highway NOA to deliver more robust and more flexible driving performance, and can handle the complex scenarios that trunk-line drivers face daily. In the future, Zhuoyu will also extend the relevant capabilities to scenarios where truck drivers commonly travel, such as national roads and provincial roads, to achieve more complete trunk-line network coverage.

Debuting Jimu 2.0, enabling heavy trucks to see far, see accurately, and see clearly

Panxing's safety capability comes not only from the "brain" but also from a good pair of "eyes."

At the launch event, Zhuoyu officially disclosed that Panxing will be the first to carry the Jimu 2.0 system. Compared with traditional split-type LiDAR solutions, Jimu 2.0 is installed inside the cabin, so it is not exposed to rain or snow, cannot be struck by flying stones, and requires no separate cleaning device, making it more suitable for the long-term, high-intensity operating environment of commercial heavy trucks.

At highway speeds, Jimu 2.0 can detect up to 500 meters; it can identify small, irregularly shaped, and low-reflection obstacles in the distance, such as black tires and fallen cardboard boxes; and in scenarios involving strong light, rain and fog, or dirt, it can also filter noise through algorithms to achieve more reliable all-scenario perception performance.

A full-dimensional safety system and fine-tuning of heavy truck dynamics improve long-term operational stability

Panxing's safety is not a configuration table but a complete system.

The first layer is multi-scenario safety reminders, which can issue early warnings for risks such as speeding on curves, pedestrians crossing, and lateral obstacles; the second layer is all-weather active safety, where the system can actively avoid danger at critical moments if the driver does not intervene; the third layer is multiple safety redundancies, including fault self-diagnosis and minimal risk strategies, so that even if a problem occurs, the vehicle can execute safer handling actions.

In terms of vehicle control, heavy trucks have their own unique dynamic characteristics. A 17-meter trailer may experience "sway" when turning or changing lanes, and combined with inner-wheel difference and blind spots, this brings higher control difficulty. Zhuoyu Panxing uses vision to identify trailer posture in real time, estimate the trailer articulation angle, and combine this with real-time load perception to dynamically predict collision risks.

For scenarios such as uphill and downhill sections, continuous large curves, and rough roads, the system has also been finely tuned for dynamics. Behind every technical detail is an engineering embodiment oriented toward mass production and real operations.

The world's first AI reinforcement learning fuel-saving algorithm, with measured fuel savings of 4.5%–7.5%

Saving money is one of the most practical needs of commercial vehicle customers.

In heavy truck operations, fuel costs account for 30% to 40% of operating costs. Experienced drivers can achieve lower fuel consumption by anticipating road conditions and reasonably using retarders, auxiliary braking, and foot brakes. Zhuoyu hopes to use AI to learn from and surpass this experience.

At the launch event, Zhuoyu introduced the world's first AI reinforcement learning fuel-saving algorithm. Through simulation training, the algorithm enables the model to "drive by itself, test by itself, and figure things out by itself" in complex environments, comprehensively considering a large number of variables such as slope, curvature, transmission system, braking system status, and surrounding vehicle conditions, accurately calculating the golden fuel-release point and golden fuel-supply point, and intelligently scheduling gears and multiple braking systems like an experienced driver.

According to Zhuoyu's statistics, over the past two months, Zhuoyu tracked more than 200 mass-produced vehicles for a cumulative total of over 20 million kilometers. After being equipped with the AI fuel-saving algorithm, vehicle fuel consumption continued to decline, with measured fuel savings reaching 4.5%–7.5%. Based on current fuel prices, this can save about 10 yuan per 100 kilometers; driving 800 kilometers a day can save about 80 yuan, and over a year it is expected to save 20,000 to 30,000 yuan.

Liu Sikang said that 20,000 to 30,000 yuan a year is profit for the fleet and real income for drivers. In addition to fuel savings, drivers are more relaxed and human efficiency improves; and after safety capabilities reduce accident risks, insurance costs are also expected to decline further.

Comprehensive estimates show that a set of intelligent assisted driving systems can pay for itself within a year. In the past, the industry often regarded intelligent assisted driving systems as a cost item; in the commercial vehicle sector, however, they can already become a tool for saving money or even making money.

Toward unmanned logistics vehicles and native multimodal foundation models, Zhuoyu continues to expand the boundaries of intelligent mobility

Beyond the mass-production launch of Panxing, Zhuoyu also announced its next-stage technology layout.

The L4 unmanned logistics vehicle business adjacent to commercial vehicle scenarios is also advancing rapidly. ROBOVAN, developed by Zhuoyu in cooperation with SAIC Maxus and Yisen Power, has entered trial operation, and cooperation with several other ecosystem partners is also actively progressing. From trunk-line logistics to urban distribution and then to last-mile delivery, Zhuoyu's intelligent footprint in the logistics industry will continue to expand.

In terms of native multimodal foundation models, Zhuoyu stated that one set of models can solve the problems of multiple sensor configurations and cross-carrier deployment. Passenger vehicles, commercial heavy trucks, buses, overseas models, and other carriers differ significantly in size, form, sensor installation positions, and driving methods. How to enable the same model to accurately understand the environment and adapt to the "driving style" of different carriers is a key challenge for cross-vertical deployment.

To this end, Zhuoyu is using the native multimodal foundation model MFM to solve the problems of multiple sensor configurations and cross-carrier adaptation, with the goal of truly achieving unified intelligent capability across verticals, regions, and scenarios, so that one "AI brain" can serve all-scenario mobility.

Conclusion

In the future, Zhuoyu will continue to promote technology coverage in more scenarios, covering both "human mobility" and "goods movement," continuously practicing the strategy of "intelligence for all mobility," and providing the world with safe and relaxed mobile intelligence.

Source: Submitted article

Images in this article are from: 123RF licensed image library. Infringement report. Back to share. Related articles. All comments.