Written by: Li He
Posted on: November 29, 2018 | 
| 中文
A schematic of JD Logistics
Recently, JD, one of the largest e-commerce and logistics companies in China and the world, has announced the success of its research and development of unmanned distribution station, and has declared that it is the world's first unmanned distribution + self-pickup site. Suning Logistics said that by 2020, it will have popularized piloted driving in terminal distribution and scaled use of unmanned delivery vehicles. Moreover, Cainiao Network points out that its unmanned warehouses, driverless vehicles and drone technology are already mature, and unmanned distribution over the whole link has been successfully tested in a small scale. The era of unmanned services by using AI, seems to be arriving first in logistics and then in express delivery industry.
So how long will it take until the technology of unmanned logistics is realized? What automatic equipment is needed? And how long can the unmanned distribution be accessible to our daily life?
Industry Layout Achieving Seamless Connection of Intelligent Logistics Network
Auto-navigated unmanned vehicles dock with the express delivery tower known as “Optimus Prime,” after running safely all the way; unmanned aerial vehicles (UAVs) which have received reservation orders on the top of the tower smoothly carry the parcels and accurately deliver them to the users’ pick-up point. Downstairs in the residential apartments sit intelligent face-scanning lockers that offer both heat-preserving and refrigerating storage, and users can open the lockers in two seconds without bringing their cell phone. Cainiao container, the magic parcel receiver at the residents’ doorstep, supports one button opening, free capacity scaling and built-in video cameras. These are the latest application scenarios of Cainiao’s unmanned logistics at Xiong’an Civic Center.
In Beijing's Huangqu Kafka Community, an unmanned yellow vehicle drove to the gate of the community and stopped to speak to the security guard, "Hello, please open the door for me." Since June, 2018, the upgraded version of the Suning unmanned delivery vehicle "Wolong-1" has started normal operation at Suning Huangqu Store in Beijing, marking the first of its kind, putting into the real-world, it’s regular use in China. From unmanned vehicles and UAVs to intelligent cabinets, Cainiao, Suning and JD are all scrambling for the early advantage of unmanned logistics to solve the last-mile problem in express delivery.
Liu Yanguang, general manager of the UAV Industry Center, under a department of JD, told Science and Technology Daily that with the application of big data, artificial intelligence, Internet of Things and other technologies, the logistics industry is gradually realizing "unmanned" operations over many links. JD has formed a new supply chain system represented by unmanned warehousing, UAVs, driverless vehicles, light trucks and unmanned distribution stations, which has been gradually applied in all aspects of logistics. Among them, unmanned warehousing addresses the issues on stock replenishment, storage, picking, and packaging and sorting. Unmanned vehicles focus on terminal delivery in the urban environment, while UAVs and unmanned distribution stations solve the last-mile delivery problem in remote mountainous areas and the countryside. Liu said that with the successful development of unmanned heavy trucks and large cargo UAVs, JD has begun to use its intelligent logistics network, integrating air and land delivery in pilot areas.
Technical Support Underpinned by Visual, INS and Other AI Hi-Techs
Chinese logistics enterprises, such as Cainiao, Suning and JD, have not only made an advance layout for unmanned logistics, but also conducted a lot of technical research and development. Zhu Lijun, Cainiao’s senior algorithm expert said, "AI is invented." Cainiao’s unmanned logistics mainly includes the digitization and linking of logistical elements, and the use of automated equipment for carrying packages.
Zhu said that logistical elements can be digitized by applying big data, cloud computing, Internet of Things and other technologies. For example, in an unmanned logistics system, if we need to know the size of a package, what commodities it contains and where it is destined, then "all of this information must be digitized before it can be read" and best matched with the vehicle’s information.
“We also use visual technology to judge the volume of goods and read data with intelligent sensors,” Zhu said. In unmanned warehouses, a large number of automated mechanical arms are used for sorting and packaging of goods, AGVs and unmanned forklifts for carrying, sensors for determining the destination of packages, and automated flow line for delivering packages to driverless vehicles.
Suning Logistics said that Wolong-1 could run 12 km/h and climb slopes of 35 degrees, with a battery life of up to eight hours. The ability of the unmanned vehicle to take to the road alone should be attributed to the support of a variety of AI high-techs, such as multi-line lidar + GPS + inertial navigation system (INS), and other multi-sensor fusion positioning methods, which enable autonomous navigation with positioning accuracy as high as 1-3 cm. In terms of perception, the fusion of lidar and visual real-time recognition technology helps the vehicle to recognize the surrounding pedestrians, cars and obstacles, thus providing the basis for planning the best detour route. Suning is also developing a “four-legged” walking Wolong-1, which will not only take the elevator, but also climb the stairs in the future. With more skills, it will provide even more intelligent services.
The driverless vehicle is moving from lightweight ones to heavy duty ones. In May 2018, a heavy-duty unmanned truck "Xinglong-1" was tested in Fengxian Park of Suning Logistics in Shanghai. The test vehicle had a load of 40 tons and used depth sensor fusion technology to realize automatic perception, cognition, decision-making and control, accurate recognition within the range of 300 meters and a response speed of 25 meters per second. Moreover, it can realize automatic emergency braking, adaptive cruise, lane deviation warning, pedestrian detection, and autonomous obstacle avoidance on the high-speed test road. With real-time kinematic survey, global positioning, real-time positioning and map building technology being combined with algorithm, its automatic driving error can be controlled within two centimeters.
"Relying on big data, AI and unmanned technology, JD intelligent logistics is a complete system," said Liu Yanguang. She explained that automation upgrade of a single link will only bring local efficiency improvement, while JD has a complete logistics system and a wide range of logistics application scenarios, which enable the company to launch comprehensive intelligence logistics upgrade, including reducing cost, enhancing efficiency and conducting intelligent transformation over each link of the whole industry chain.
Then when will the totally unmanned delivery be accessible to ordinary homes? Liu optimistically expressed that JD’s current AI equipment has already been fully put into normal operation, serving in all aspects of its intelligent logistics system. Since last year, JD’s UAVs have been used for normal operation of distribution in Jiangsu, Qinghai, Hainan and several other provinces.
"But large-scale application of unmanned logistics still faces some policy bottlenecks. For instance, there are no effective laws and regulations for unmanned vehicles on the road and UAVs in the sky. As technology updates and relevant laws improve, large-scale unmanned distribution is expected to be put into use soon,” said Zhu Lijun.
Translated by Xu Donglin
You may also like:
