A breakthrough in Australian unmanned driving technology

Australian researchers have found a way for self-driving cars to “see” the world around them. This allows you to track pedestrians hiding behind buildings and cyclists hiding in trucks and buses.

The groundbreaking development uses a so-called X-ray-style vision to penetrate beyond the blind spots of traffic.

However, unlike Superman’s ability to see through solid objects, this technology relies on multiple vehicles that share a line of sight.

The new Intelligent Transport Systems (ITS) are expected to promote a concept known as Cooperative Perception (CP).

Engineers and scientists say roadside information sharing units or ITS stations with cameras and radar sensors will allow self-driving cars to share what they “see” with others using vehicle-to-vehicle communication. I am.

This allows you to take advantage of different perspectives.

This groundbreaking achievement is the result of a three-year collaboration between the University of Sydney’s Field Robotics Center and software company Cohda Wireless.

The author believes that connecting vehicles into a single system greatly expands the collective range of perception, allowing connected cars to detect what they would not normally be able to detect.

During the test, the smart car was able to use CP information to track pedestrians visually blocked by the building, said Professor Eduardonebot, director of the Australian Field Robotics Center.

“This was achieved a few seconds before the local sensory sensor or driver could see the same pedestrian around the corner, providing extra time for the driver or navigation stack to react,” he said. Said.

“It’s a game changer for both human-operated and self-driving cars, and we hope it will significantly improve the efficiency and safety of road transport.”

Other experiments have demonstrated the ability of CP technology to safely interact with pedestrians and pedestrians rushing towards intersections.

“The connected self-driving car was able to take the preemptive action of braking and stopping in front of the pedestrian crossing based on the expected movement of the pedestrian,” Nebot said. increase.

Meanwhile, a study at the University of New South Wales has proposed the design of a highway network with lanes dedicated to self-driving cars.

Using computer modeling of mixed scenarios, engineers have discovered that dedicated lanes significantly improve the overall safety and traffic flow of hybrid networks.

Lead author Dr. Chantanu Chakraborti said the proposed model helps minimize interaction with the driver-operated vehicle and reduce overall congestion.

“Unless proper modeling is done during this transition phase, a mix of self-driving cars and legacy vehicles will cause problems with the road network,” he said.

Adding a dedicated lane would be confusing, but Chakra Volti said this was already happening on the bus.

Freeways is great for trying out ideas with dedicated entrances and exits that allow drivers to automatically turn automatic features on and off, he added.

You can also use variable signs to change lane designations based on traffic conditions.

John Kidman