China’s answer to Google’s self-driving car has set its own record on the roads of Beijing. The Baidu robot car became the first Chinese vehicle to complete a fully autonomous drive under a variety of road conditions that included navigating local road traffic and highway driving.
The Baidu autonomous vehicle—a modified BMW 3 Series sports sedan made by the Chinese search giant—reached speeds of up to 100 kilometers per hour during the course of the test drive, according to a 10 December press release. Its robotic actions included making right and left turns, U-turns, slowing down when detecting vehicles ahead, changing lanes, and passing other cars. The car also successfully merged into traffic from on-ramps and exited from off-ramps.
Baidu’s test drive followed a 30-kilometer course from the company’s Beijing headquarters near Zhongguancun Science Park, continued on a stretch of the G7 highway, ran along the Fifth Ring Road expressway, cut through Olympic Park, and ended back at Baidu headquarters.
The Baidu Institute of Deep Learning has led the company’s autonomous driving project since 2013. Its approach to developing driverless cars relies upon development of highly automated driving (HAD) maps that work in combination with other self-driving car technologies. It’s an incremental approach that relies on adaptation to different environments rather than increasing overall levels of vehicle autonomy.
For example, self-driving vehicles running the same routes again and again can gradually train the “AutoBrain" autonomous technology’s computer vision and deep learning systems to become accustomed to new environments. In that way, Baidu hopes to accomplish HAD mapping for a majority of China’s roadways within the next decade.
Baidu first publicized its mapping approach to self-driving cars during a government conference in Beijing in April. The Chinese tech giant has suggested that a self-driving car application could help attract customers to its other map-based services.
If self-driving cars take off in China, perhaps more efficient robot driving could help partially alleviate the country’s serious gridlock problems. But the country’s air pollution problem, described by some as “airpocalypse,” may require more drastic steps such as fewer cars on the roads and cleaner technologies to reduce the overall amount of car emissions.
Jeremy Hsu has been working as a science and technology journalist in New York City since 2008. He has written on subjects as diverse as supercomputing and wearable electronics for IEEE Spectrum. When he’s not trying to wrap his head around the latest quantum computing news for Spectrum, he also contributes to a variety of publications such as Scientific American, Discover, Popular Science, and others. He is a graduate of New York University’s Science, Health & Environmental Reporting Program.