To deliver food to quarantined areas UDI a Chinese startup deploys ‘running robot’ vehicles.

Self-driving delivery vehicle developed by Unity Drive Innovation (UDI)
Photo: UDIA self-driving vehicle departs a distribution center to deliver fresh fruits, vegetables, and other supplies to residents in Zibo, in the Shandong province of China

Zibo communities in eastern China province have been receiving its vegetable supply of 1,000 Kg per month delivery using a robot van. This van is the result of how technologically innovative China is today. The van is actually an invention of a Chinese startup company, UDI, and appropriately named Hercules, a demi-god of delivery vehicles. It can actually boast of a number of the most advanced technology available today

Diagram showing self-driving delivery vehicle created by Unity Drive Innovation (UDI)
Image: UDIUDI’s vehicle is equipped with a main lidar and three auxiliary lidars, a stereo camera, and various other sensors [top]. The cargo compartment can be modified based on the items to be transported and is not shown. The chassis [bottom] features a vehicle control unit (VCU), removable lithium-ion battery, motor control unit (MCU), electric power steering (EPS), electro-hydraulic brake (EHB), electronic parking brake (EPB), on-board charger (OBC), and direct-current-to-direct-current (DCDC) converter.

This includes an industrial-grade PC with a Robot Operating System, or ROS installed, a drive-by-wire, i.e. electronic systems to activate brakes, and control steering, and control electric motors, run by an 8.4-kWh lithium-ion battery. Hecules uses 4 lidars (Light Detection and Ranging technology much like the radar only this uses eye-safe laser light), one main at the rooftop and three auxiliary lidars to “see” and accurately gauge the distance of objects in its surroundings. For another sense, it uses a stereo camera, four fisheye cameras, 16 sonars. For localizing its location in the stored map data redundant satellite navigation systems, an inertial measurement unit (IMU) used in unmanned aerial vehicles (UAVs), and two-wheel encoders to determine vehicle velocity and acceleration.

Artificial Intelligence drives the vehicle for truly autonomous driving

RoboVan has 3 AI algorithms or learning programs. These are the main perception algorithm, which consists of a convolutional neural network and trained to detect and classify objects and compares the received lidar point-data from the sensors the PC received. This results in a set of the neural output of 3D bounding boxes representing vehicles and other obstacles on the road. This process is repeated or cycled 100 times per second for high accuracy even when the vehicle is in motion.

The second algorithm brings the images from forward-facing cameras to identify road signs and traffic lights, and the third algorithm allows the vehicle to self-localize by processing the matches the point-data and IMU data to a global map.

The PC sends commands to two secondary computers running real-time operating systems and connected to the drive-by-wire modules to accelerate, brake, and steer and essentially making unmanned driving possible.

Professor Ming Liu, a computer scientist at the Hong Kong University of Science and Technology (HKUST) and cofounder of Unity Drive Innovation, or UDI, the Shenzhen-based startup that developed the self-driving van said “The unmanned vehicle provides a ‘contactless’ alternative to regular deliveries, helping reduce the risk of person-to-person infection.”

Despite the lockdown often encountering busy traffic conditions the vans have made more than 2,500 autonomous trips since February, UDI has been operating a fleet of vehicles in Zibo and Suzhou and Shenzhen, where they deliver vegetables to communities and meal boxes to checkpoint workers and spray disinfectant near hospitals.

Autonomous vehicle developed by Unity Drive Innovation (UDI) sprays disinfectant near a hospital in Shenzhen, China, during the COVID-19 pandemic
Photo: UDIOne of UDI’s autonomous vehicles equipped with a device that sprays disinfectant operates near a hospital in Shenzhen.

Professor Liu told IEEE Spectrum in an interview via Zoom “It’s like Uber for packages—you use your phone to call a robot to pick up and deliver your boxes,”

Self-driving vehicle developed by Unity Drive Innovation (UDI) delivers meals to checkpoint workers
Photo: UDIA self-driving vehicle delivers lunch boxes to workers in Pingshan District in Shenzhen. Since February, UDI’s autonomous fleet has made more than 800 meal deliveries.

Professor Liu says UDI faces more challenging driving conditions in Shenzhen, for example, the UDI vans have to navigate through narrow streets with double parked cars and aggressive motorcycles that whiz by narrowly missing the robot.

UDI has monitored its fleet from its headquarters for over the past couple of months using 5G, with just 10 milliseconds of delay a remote operator can receive data from a vehicle. When the robot vans in Shenzhen encountered situations they lacked learning data like false detections of traffic lights, too many vehicles on the road, human intervention was required in about 24 instances.

Professor Liu says it’s a challenge to balance cautiousness and aggressiveness in self-driving vehicles that will operate in the real world. He notes that UDI vehicles, or what he likes to call “running robots,” have been collecting huge amounts of video and sensor data during their autonomous runs. UDI plans to open source part of the data. This information will be useful to improve computer simulations of the robot vehicles and, later, the vehicles themselves.

RoboVans mass production

Cofounders of UDI are Professor Xiaorui Zhu at Harbin Institute of Technology, in Shenzhen, and Professor Lujia Wang at the Shenzhen Institutes of Advanced Technology, part of the Chinese Academy of Sciences, “We want to be the first company in the world to achieve mass production of autonomous logistics vehicles,” says Wang, who is the CTO of UDI. Hiring 100 employees the startup is preparing in the next several months to put its assembly line into high gear.

“I’m not saying we solved all the problems,” Professor Liu says, citing system integration and cost as the biggest challenges. “Can we do better? Yes, it can always be better.”

(Source)

Jonathan Morales

By Jonathan Morales

Likes to write technical pieces and read cutting edge tech news. Watch Action Kodi Videos and general tinkering and repairing home gadgets.