CSIRO autonomous car breakthrough: bionic eye tech could help future cars ‘see’

AN Australian bionic eye invention could hold the key to making autonomous cars navigate city driving scenarios.

AUSTRALIAN technology used in bionic eyes — that improve sight for humans with vision impairment — could be the breakthrough that enables autonomous cars of the future to detect tricky traffic situations, the CSIRO claims.

The CSIRO’s Data 61 division is working with a technology company that supplies Chinese car makers to adapt its bionic eye know-how to help navigate dangers on the road.

As car makers around the world roll out their autonomous tech, most are starting with “on-ramp to off-ramp” freeway automation, because there are fewer obstacles to detect. Most vehicles are travelling at a similar speed, the lane markings are typically good, and there are no pedestrians.

The bigger challenge for autonomous cars is reading traffic signals, stop signs, pedestrians and cyclists and other random obstacles in cities and suburbs.

The CSIRO’s Data 61 division is working with China’s ZongMu Technology to equip vehicles with computer vision “to allow a machine to see and understand the environment the way humans do, and react to hazards”.

Dr Nick Barnes (left) and Dr Shaodi You (right) from the CSIRO’s Data 61 division are testing the suitability of bionic eye tech for autonomous cars. Picture: Supplied.

In essence the scientists are developing “algorithms to estimate the space between objects according to the vehicle’s motion, and predict the potential hazards of moving objects”.

“Computer vision is the technology that allows autonomous vehicles to determine the difference between what is pavement and what is driveable road,” said CSIRO Data 61’s Dr Nick Barnes.

“Unlike laser sensors which rely on a series of points to identify hazards, computer vision offers richer information and a deeper understanding of road scenes through 3D image analysis, enabling safer automated driving.”

Traffic and pedestrians are bigger challenges for autonomous cars than freeway driving, as this Ford test demonstrates. Picture: Supplied.

Senior Research Scientist at CSIRO Data 61, Dr Shaodi You, said the technology will allow autonomous vehicles to quickly react to “any hazards” from 10 metres or further to avoid collisions.

“Our technology will allow self-driving cars to more quickly detect and avoid hazards, understand and obey road rules and to determine their exact location in relation to other moving vehicles and landmarks in a given environment,” said Dr You.

“The laser sensors used by the majority of companies are prohibitively expensive. On the other hand, the computer vision algorithms we’re developing with ZongMu cost one-tenth the amount and will allow commercial and truly autonomous cars to reach the road in a much shorter time.”

However, CSIRO Data 61 is yet to put its theory into practice in a car on real roads.

“The underlying computer vision technology has been proven to work in past projects (such as) the bionic eye,” said CSIRO Data 61 representative, Chris Chelvan.

The technology is expected to be tested in a trial car in China within a year.

Car giants Ford and General Motors have announced they will have commercially available, fully autonomous cars on the road by 2021. Other brands are set to follow.

Ford tested an autonomous car with Dominos pizza in the US in August 2017. The only hurdle: getting the pizza from the car to the customer’s front door. Picture: Supplied.

“The (CSIRO Data 61) technology is a work in progress, but provides a unique solution in the autonomous vehicle market,” said Mr Chelvan.

Shanghai-based ZongMu is a supplier of driver assistance systems to the Chinese car industry.

Among the automotive brands listed as ZongMu customers are FAW and GAC, both emerging car makers established just 21 years ago.

Adrian Turner, CEO at the CSIRO’s Data 61 division, said the Chinese partnership would accelerate the arrival of “commercially viable self-driving cars”.

“This partnership is building on our previous work in developing a bionic eye, using computer vision that has given sight to the visually impaired,” said Mr Turner.

This reporter is on Twitter: @JoshuaDowling