Author | Zhou Zhiyu
The Robotaxi (autonomous driving taxi) industry is entering a brutal &#34;Warring States period.&#34; In the past, companies were willing to invest in the technological stories of autonomous driving and the tackling of long-tail scenarios; now, with major players going public one after another, achieving scale and profitability has become the focus of competition.
On March 9, Geely Farizon New Energy Commercial Vehicle Group signed a deepened strategic cooperation agreement with WeRide, announcing that the all-new pre-production Robotaxi GXR will roll off the production line in the third quarter of this year, with 2,000 units to be delivered within the year. The logic of the production line is also changing: the production cycle per vehicle has been violently compressed from 1 hour in the retrofitting mode to under 10 minutes.
Prior to this, the Robotaxi industry generally faced a deadlock of &#34;difficult mass production and high costs.&#34; With the breakthrough by Geely Farizon and WeRide, the industry has officially crossed the initial stage of making prototype vehicles. A fierce battle over capacity, costs, and delivery has thus begun.
In this process, whoever can first establish a closed loop for large-scale implementation will survive the earnings season in the capital market, and whoever can do so will be the first to reach the shore in this elimination competition.
<h2>Breakthrough</h2>
In the past few years, the Robotaxi sector has been bustling, but it has never been able to shake off the awkward label of being a retrofitting factory.
Buying a mass-produced vehicle, disassembling, drilling, and adding radars in a retrofitting workshop, this model inevitably leads to poor consistency and high costs, becoming the biggest obstacle to commercialization. For Han Xu, founder and CEO of WeRide, this pain is tangible: &#34;A car scattered there, a bunch of people rushing to install it, that’s not called mass production; that’s called a workshop.&#34;
Pre-production has become the tool for WeRide to break through its current development bottleneck.
A core aspect of this press conference is that the Robotaxi GXR has achieved a truly &#34;native&#34; manufacturing process. Relying on Geely Farizon's mature supply chain system and drive-by-wire AI chassis technology, the GXR has taken into account the hardware requirements for autonomous driving from the very beginning of vehicle design (SOP stage).
Geely Farizon CEO Fan Xianjun made an analogy to Wall Street News, stating that previously, the front and rear installations were separate, requiring two tests and two integrations. Now, with pre-production achieving integration, it is as smooth as a person's native organs. If a person has a prosthetic limb, you need to re-integrate and re-recognize it; but if it is original, it does not require that.
The direct benefit brought by this &#34;native&#34; approach is a significant reduction in costs. Specifically, this industrialized dimensionality reduction directly lowers the manufacturing cost of the vehicle by 15%. In the single vehicle economic model (UE) of Robotaxi, vehicle depreciation is the largest explicit cost, and this reduction is sufficient to bring the breakeven point forward. Additionally, through pre-production integration optimization, the power consumption of the autonomous driving kit has been reduced by 50%, directly cutting daily operational energy expenses.
Han Xu calculated a figure: &#34;Previously, we made retrofitted vehicles, which were modified by enthusiasts. Now, Geely Farizon's production line has shortened the production cycle per vehicle from 1 hour to under 10 minutes. This is a hallmark of modern industrial civilization—mechanization and mass production.&#34;
In addition to the qualitative change in manufacturing efficiency, the perception capability has also simultaneously leaped forward, no longer merely a stacking of parameters
The GXR is equipped with WeRide's latest Gen8 autonomous driving suite. Its core thousand-line main lidar enhances point cloud resolution by 17 times, extending the detection range to 600 meters, which is 2-3 times that of mainstream industry solutions.
This is not just for show. In high-speed scenarios where commercial vehicles carry dozens of tons, the braking distance can be infinitely extended by physical laws. Han Xu explained this technological upgrade, stating that the vehicle can see 2-3 times farther than others, allowing it to gain over 70% of reaction time margin in high-speed scenarios.
This redundancy is designed to ensure all-weather operational capability in extreme weather conditions such as heavy rain and dense fog—after all, true commercial operations do not discriminate against weather.
A deeper breakthrough lies in the redefinition of &#34;safety.&#34;
Fan Xianjun revealed a detail: for one parameter, he and the product manager debated for 40 minutes. Ultimately, this B-pillar-less model passed rigorous testing—endurance pressure of 45,000N and body torsional rigidity reaching 38,000N·m/deg, far exceeding safety standards.
Additionally, targeted engineering optimizations were made for common &#34;minor bumps&#34; during Robotaxi operations. Han Xu mentioned that previous vehicles used a line sensing method, which might not recognize collisions that were slightly higher or lower (such as helmet impacts). Now, the body sensors have achieved full wrap-around surface sensing, capable of sensitively detecting collisions while filtering out false alarms from road debris.
This comprehensive reconstruction from micro sensors to macro body structure marks that Robotaxi has finally shed its patchwork feel. It has become an industrial standard product born for autonomous driving.
For the industry, this means the dimensions of competition have changed: it is no longer about whose demo runs smoother, but rather who can industrially produce Robotaxis at a lower BOM cost and higher yield rate.
<h2>Change</h2>
A peak battle for the commercialization of Robotaxis has already entered a climax with the establishment of the &#34;iron triangle&#34; model.
From a market perspective, although domestic Robotaxis have been piloted in many locations, they face fierce competition from the declining prices of high-end intelligent driving for C-end consumers. As Tesla's FSD and Huawei's ADS attempt to turn passenger cars into part-time Robotaxis, dedicated L4-level operational vehicles must seek a wider moat.
Going overseas to find a &#34;new continent&#34; with stronger payment capabilities and broader market space has become a consensus among leading players. However, going abroad is not easy, especially in the current context of complex geopolitical situations and stringent data security reviews.
Geely's Long-distance and WeRide have explored an &#34;AI technology (WeRide) + vehicle manufacturing (Long-distance) + global operations (scenarios)&#34; iron triangle path. This is not only an innovation in business models but also a survival wisdom amid geopolitical games.
Geely's Long-distance provides not only production capacity but also a &#34;compliance passport&#34; to the global market. The prototype of the GXR, the &#34;Long-distance Super VAN,&#34; has already entered nearly 30 European markets.
After the collaboration, WeRide no longer needs to struggle to prove the basic safety of its vehicles to European regulatory agencies, as Long-distance has already stood in front with the most stringent NCAP five-star certification
Han Xu expressed a mix of helplessness and confidence as a Chinese technology company when discussing the pain points of overseas operations: &#34;When we go to the Middle East, the endless yellow sand makes the moisture in our skin evaporate. When the vehicles arrive there, aging seals and door noises are common. But the remote vehicles have never had any issues.&#34;
He even bluntly stated: &#34;If it weren't for geopolitical factors and various trade protectionism, I believe that with the cost-performance ratio of this vehicle, sweeping the European market would be no problem at all. That vehicle (the remote super VAN) is much better compared to European competitors.&#34;
Currently, both parties have achieved operations in Abu Dhabi and Dubai, and will officially open in Singapore on April 1 of this year. This capability of compliance in advance gives WeRide a wider moat in dealing with overseas regulations compared to pure technology service providers.
Another reality is that the future commercial Robotaxi scenario will be more complex, covering long-distance transportation and cold chain logistics. In this field, pure electric solutions are not a panacea. Fan Xianjun also suggested that methanol is one of the solutions.
&#34;As the penetration rate of new energy increases, the scenarios for commercial vehicles will extend from urban delivery to medium and long-distance transportation,&#34; Fan Xianjun analyzed. &#34;In this field, pure electric cannot solve the endurance pain points, hydrogen is too expensive, while methanol, as 'liquid hydrogen', is the best solution for energy independence and cost.&#34;
This strategy has extremely high strategic value for Robotaxi and even future Robotrucks (unmanned trucks). Current Robotaxis are mostly modified from pure electric models, which face significant range anxiety in low-temperature or long-distance scenarios. The &#34;methanol electric + pure electric&#34; dual technology route planned by WeRide provides an energy interface for autonomous driving to land in more complex commercial scenarios. This is also why Han Xu said: &#34;Wherever a person can drive, we hope to have a set of universal technology to do it. In the future, no matter what commercial vehicle WeRide makes, its autonomous driving technology should ideally be provided by WeRide.&#34;
Industry expectations are that, driven by Geely's WeRide and WeRide, the Robotaxi industry chain will undergo a reshuffle. Players who cannot achieve pre-installed mass production or cannot lower hardware costs will lose competitiveness in the face of industrialization's relentless wheels.
Riding the wave of pre-installed mass production, WeRide's fleet scale will exceed 2,600 vehicles this year. This is just the beginning.
Han Xu made an aggressive prediction on-site: the growth of Robotaxi will follow a curve faster than Moore's Law, with integration doubling every 18 months, prices halving, and quantities increasing by 2.5 to 3 times each year. By 2030, the fleet size will reach tens of thousands; by 2035, this number will be 1 million.
In 2026, when the Robotaxi industry collectively enters the &#34;post-IPO era,&#34; the patience of the capital market will have run out. Whoever can first establish a closed loop for large-scale landing, and who can make high technology affordable through industrialization, will be able to write a beautiful growth curve in their financial reports.
For those players who still cannot solve the problem of pre-installed mass production and are still tightening screws in the retrofitting workshop, their time is running out.
The second half of Robotaxi has no warmth, only the survival of the fittest

PONY

00175

The major test after capitalization

- The Robotaxi industry faces intense competition as major players focus on achieving scale and profitability following their public listings.  
- Geely's commercial vehicle group and WeRide have announced a strategic partnership, aiming to deliver 2,000 new Robotaxi GXR units this year, significantly reducing production time.  
- As the industry evolves, those who can successfully implement mass production and lower costs will dominate, while inefficiencies could lead to the exit of less competitive players.

Robotaxi enters the "ten-minute" mass production era