Auto - Show Biz clinic originally published at Auto - Show Biz clinic
10 (China Daily), 08;29, February 1, 2023
1010Photo of an in-service auto driving robotaxi from Baidu’s Apollo Go, southwest China’s Chongqing municipal, Aug. 13, 2022. (People’s Daily Online/Rong Zengfu)
Sensors and software becoming essential components of new vehicles
Sun Congjun’s 2012 Mercedes C200 was replaced by an electric Xpeng last summer.
He is particularly attracted to the automatic parking that Nvidia chips, four cameras, and 12 ultrasonic sensor enable, among other things.
The veteran driver stated that he is good at reverse but feels better when the car takes control — looking for a parking space and reversing into one — so he can grab his phone and briefcase and be ready for exit.
“It’s not about time, but it’s nice to enjoy the technology can do for us, isn’t it?” He said.
Sun is not the only one. According to the Ministry of Industry and Information Technology, 2.88 million passenger cars with driver-assist systems were sold in China during the first half of the last year. They accounted for 32.4 percent of all passenger vehicles sold in the nation during that period, with sales up 46.2 percent year-on-year.
Edward Wang, managing director for syndicated research at J.D. Power China, which focuses primarily on consumer insights, data, and analytics, stated that smart features and amazing onboard experiences are becoming more important factors in people’s decision to buy a car.
He said, “They aren’t necessarily buying a model solely to have those functions but they sure wouldn’t buy one if it did not have such functions.”
Level 2 is for driver-assistance systems. This is one of six levels of driving automation that the Society of Automotive Engineers has. Level 0 means no automation and Level 5 means that the car does not require a driver.
Other than parking, Level 2 functions include acceleration and deceleration based traffic flow, lane departure warnings, and autonomous emergency brake systems.
These functions are a must-have in China’s new models, which include compact cars starting at less than 100,000 Yuan ($14,370).
Xu Changming is the deputy director of China’s State Information Center. He stated that Generation Z will drive China’s development of smart vehicles. “They have grown up using artificial intelligence and the internet, so they like new technology,” said Xu. He estimated that about 35 percent of China’s car buyers will be born after 1995.
Development moves
The carmakers are also increasing their efforts to develop these capabilities and partnering up with large-name technology companies.
Tencent, a tech giant, and electric car startup Nio, joined forces in November to develop high-precision mapping systems that can be used by drivers. The deal will allow Nio access to Tencent’s cloud computing infrastructure to store data and train its driver-assistance technology.
After Xpeng announced in August that it had opened a computing center to train software for its smart cars, Alibaba and Xpeng formed the partnership. Xpeng stated that the center will reduce the time required to train its core autonomous driving model from seven to one hour. Alibaba’s cloud division will also be used by the startup for its computing needs.
He Xiaopeng (Xpeng’s CEO), a former Alibaba executive, stated that “We will need at least 1 million yuan per year on computing power by 2025.” “[Photo/China Daily]
Nvidia’s Drive Orin chip is the world’s most powerful autonomous driving processor. It first appeared in Nio’s ET7 sedan last Year.
William Li, founder of Nio, stated, “Our in-house group has developed autonomous driving algorithms that run on four industry-leading Nvidia Orin processings, delivering an unprecedented 1,000 TOPS (teraoperations per second) in production vehicles. “
It is expensive to develop an autonomous driving system. Some carmakers purchase the driver-assistance system directly from suppliers.
Haomo.ai is an autonomous driving startup that was supported by China’s Great Wall Motors. It said that its driver-assistance software HPilot has been installed on about 20 Chinese models, with the total distance covered by the system exceeding 23,000,000 kilometers.
The HPilot system is used in the Wey Mocha, a SUV made by Great Wall Motors. The model can spot traffic lights and change lanes, as well as automatically park.
It is China’s first mass-produced vehicle to use Qualcomm’s Snapdragon Rider, the chipmaker’s platform for autonomous driving functions. The model also includes 31 sensors, including two lidar sensors.
QiaoXinyu, Wey’s deputy general manager, stated that smart driving will be the decisive element in the highly competitive new-energy vehicle segment.
Zhang Kai, Haomo chairman, stated that smart vehicles with advanced driver-assist functions will make up 70% of all new Chinese vehicles by 2025.
Volkswagen, a German carmaker, has established the first Chinese subsidiary of its software arm Caridad. This software arm develops software for everything from infotainment and autonomous driving.
Cariad’s CEO Dirk Hilgenberg stated in an interview with China Daily that “as we all know, China is very progressive in automated driving, and digital experience. We want to make sure we can provide excellent service to local customers by establishing a subsidiary company that is able grasp the speed, development, and innovation in China. “
Cariad decided to partner with Horizon Robotics, a Chinese startup that provides computing solutions for smart cars. The investment is 1.3 billion euros ($1.35 million).
Cariad will own a 60% share of the joint venture that will create driver-assistance, autonomous driving solutions for its electric cars in China.
“The China market has so much potential and customer expectations are high.” Hilgenberg said. Hilgenberg stated that we need to accelerate our pace for innovation.
Yuan Shengyue is a smart-vehicle expert at BAIC Bluepark. He said that China is one of the top countries for introducing Level 2 functions to production vehicles.
Yuan stated that “Level 2 functions are commonplace” and that many carmakers are working to create Level 3 and higher functions.
Mercedes-Benz, a German carmaker, is currently in discussions with Chinese authorities to introduce Level 3. Level 3 vehicles, also known as “conditional driving automation”, can drive themselves in certain conditions so that the driver is not required to monitor road conditions.
Honda, a Japanese carmaker, stated that it would develop technology to enable Level 3 self-driving capabilities by the second half the decade.
Mahito Shikama, an engineer at Honda, told reporters during an online briefing that he believes he can reduce accidents by pursuing Level 3 technology. This is in essence about cars that don’t crash. “
The Japanese company aims to achieve a zero-fatality rate in traffic accidents involving its cars or motorcycles worldwide by 2050.
.
Andrew Grant, senior associate at Bloomberg NEF’s Intelligent Mobility group, stated that Level 2 vehicles will account for more than half of all global passenger vehicle sales by 2025. He predicted that the peak level 2 sales would be in 2028 with 53 million vehicles being sold.
Robotaxi services
A Baidu robotaxi driving on a street of Wuhan.
Robotaxi services
Driverless technologies are used primarily in robotaxis and autonomous buses and trucks in designated areas. They are not private cars that can travel anywhere and may encounter unforeseen circumstances.
Baidu, the technology giant, announced in November that it would establish the largest autonomous ride-hailing area in the world as soon as this year.
Apollo, the company’s project, has already launched fully driverless ride-hailing services to Wuhan, capital of Hubei Province, and Chongqing.
.
It also has fleets with operators in vehicles in about 10 other Chinese cities.
Apollo completed 474,000 rides in the third quarter of the last year, an increase of 311 percent over the previous year and a 65 per cent increase over the prior quarter.
Chen Jingkai is an autonomous driving expert at Baidu Apollo. He said that “Baidu Apollo created a safe and intelligent autonomous driving technology system that allows us to bring robotaxi services to the open road from designated areas.” “
Baidu claims it can launch its robotaxi fleets into a new city in as little as 20 days. Its automated systems generate 96 per cent of the maps required, which reduces the work for human engineers.
Other companies, including Pony.ai, WeRide and Didi Chuxing, also operate autonomous buses and robotaxi fleets in cities from Beijing to Guangzhou, both in Guangdong.
Japanese carmaker Nissan announced in November that it would establish a robotaxi fleet at Suzhou, Jiangsu province. WeRide will provide technical support.
Hideki Kimata is the president of Nissan China Investment. He stated, “Through this new company, we aim to provide all-new riding experiences and easier, more convenient mobility service for more customers in China.” “
According to IHS Markit, China’s market for robotaxi services will surpass 1.3 trillion Yuan by 2030. This represents 60% of China’s ride-hailing market.
BloombergNEF projects that China will have the largest fleet of robotaxis in the world, with approximately 12 million vehicles by 2040. The United States will follow with around 7,000,000 autonomous vehicles.
Robotaxis is not the only option. Highly automated buses and trucks are also being seen in designated areas of China, such as ports or industrial parks.
Inceptio Technology, an autonomous driving startup in China, stated that its fleet of trucks had traveled more than 6,000,000 km.
The company was established in Shanghai in 2018 and specializes in Level 3 and 4 autonomous technologies in trucks. Two models, developed in collaboration with Sinotruk and Dongfeng, were put into production in the latter half of last year.
The startup stated that it is updating the system, making breakthroughs in its core technology including algorithms and data.[Photo provided to chinadaily.com.cn]
Yu Qian, CEO and founder of QCraft, a Chinese autonomous driving startup, said the sector is arriving at a “golden inflection point of growth”, with policies becoming increasingly clear and open, and technologies such as 5G and algorithms becoming more powerful.
Updating policies
China is one of the countries that are updating their policies quickly to support the technology’s advancement. Over 30 cities and provinces have adopted regulations that govern road testing for smart cars.
Xin Guobin (Vice-Minister of Industry and Information Technology) stated that China will introduce more policies and intensify international exchanges to create a better working environment for its burgeoning sector.
The ministry released a draft guideline in November that aims to expand road tests for autonomous driving cars nationwide. Beijing also offered permits to allow companies to move safety operators from behind their steering wheel to the back of their test vehicles in designated areas in Beijing.
Experts said that although human operators are still needed, this move is a significant step towards the commercialization fully self-driving technology.
Auto suppliers are also working hard in order to meet the demand for autonomous vehicles.
Gaogong’s statistics show that more than 74,000 lidar sensors were installed in the first ten months of the year. This is compared to less than 10,000 in the same time period in 2021.
.
Gaogong expects the figure to reach 400,000 to 500,000 units in China this year. Gaogong anticipates that the number will reach 400,000 to 500,000.
ZF, a German supplier of auto parts, announced in August that it would invest 320 million Yuan in Shanghai to expand its plant to produce electric steering systems. The company claimed that the system can be used to support autonomous driving functions at different levels.
Goodyear Tiremaker stated that it has been working closely with autonomous driving companies like TuSimple to provide tires and management solutions.
Pietro Saletta is vice-president, Goodyear Asia-Pacific consumer business. “
(Web editor: Cai Haaruo, Liang Jun).
Auto - Show Biz clinic originally published at Auto - Show Biz clinic