Alibaba Cloud, the cloud computing arm of Alibaba Group, announced today at its Alibaba Computing Conference 2016 that it will apply its artificial intelligence (AI), deep learning and data analytics capabilities to two new developments in China: smart traffic system and astronomical big data analysis.
Running at the seventh consecutive year, the four-day Alibaba Computing Conference 2016 was held at the International Convention Center in Yunqi Cloud Town in Hangzhou, China. The event attracted 40,000 participants, and some 6.5 million viewers who viewed the video feeds online.
Hangzhou City Brain automates traffic signals
"The world's longest distance is not between the South Pole and North Pole, or between friends that dine together but keep staring at their own smartphones.
"The world's longest distance is between a city's surveillance cameras and the traffic lights, as they sit on the same pole, but were never connected with each other through data," said Jian Wang, chairman of Alibaba Group Technology Steering Committee.
"To covert an image that is taken by a traffic surveillance camera into the data required to direct traffic light signals, you will likely need the data to pass through several grids, several government departments, and a number of enterprises," he continued.
Wang added that Hangzhou city is now exploring the possibility of having its roadside surveillance cameras to direct the traffic light signals.
Initiated by the Hangzhou government, the "Hangzhou City Brain" is designed to address urban challenges, primarily, the traffic congestion problem.
The "Hangzhou City Brain" will serve as a hub to consolidate data and provide real-time analytics. It relies on Alibaba Cloud's A.I. program "ET," as well as big data analytics capabilities to perform real-time traffic predictions, using its video and image recognition technologies.
The backbone of the "Hangzhou City Brain" is Apsara, Alibaba Cloud's large scale computing operating system. Apsara is able to cluster millions of servers into a super computer and to support a multitude of cloud-based services by analyzing terabytes of data points. Using propriety algorithms, this computational engine is said to be one of the largest of its kind in the world.
With the use of Apsara, Hangzhou City Brain can "view" the videos taken by some 50,000 roadside surveillance video cameras in Hangzhou within a short time frame, and make [traffic signal] decisions in just seconds.
For example, the automated traffic system capabilities can make intelligent adjustments of traffic lights on the spot: When a vehicle changes direction, the green light will automatically be extended. The pilot of the world's most advanced smart traffic management system in the Hangzhou Xiaoshan District, which commenced last month in September, has seen an increase in traffic speed by 11%.
The "Hangzhou City Brain" will also support the city's transportation departments to ease traffic congestion, and provide users with real-time traffic recommendations and travel routes.
The "Hangzhou City Brain" is being led by the Hangzhou government in coordination with 13 firms, including Alibaba Cloud, Foxconn, H3C. A research and development team of scientists from various companies has also been formed as part of the project.