New technology gives autonomous vehicles ‘Xray’ vision to aid them in tracking pedestrians, cyclists, or other vehicles that might be obscured.
Experts in Australia are now attempting to commercialize the technology, known as cooperative or collaborative perception (CP).
It involves the installation of roadside information-sharing units (‘ITS stations’) equipped with sensors such as cameras and lidar.
For example, at a busy intersection, vehicles might use these units to share what’s’seem’ with other vehicles.
This gives each vehicle an X-ray-style vision that can see through buses to see pedestrians and a fast-moving van around corners that are about to run red lights.
Example of a CP scenario at intersection. The car on the left would be in a position to alert the other vehicle of what’s happening, namely that a pedestrian crosses the road.
Artificial intelligence (AI) powers autonomous vehicles. It’s trained to recognize pedestrians so that it can stop and avoid collisions.
But they cannot be widely adopted until they are more safe than human drivers.
It is therefore crucial that they learn how to respond to different situations with the same ability as humans to ensure their full integration.
iMOVE is a government-funded research center that is executing the Australian project. With support from Cohda Wireless (transport software firm) and the University of Sydney
After three years of research, they have released their final report.
The technology’s applications are now being commercialised by Cohda, following the R&D work, which involved trials on public roads in Sydney.
Professor Eduardo Nebot, from the University of Sydney’s Australian Centre for Field Robotics, said that “This is a game-changer for both human-operated as well as autonomous vehicles” and that it will significantly improve safety and efficiency of road transport.
‘CP enables smart cars to break the practical and physical limitations of onboard perception sensor’
A vehicle equipped with the tech was capable of following a pedestrian who was visually blocked by a structure in one test.
This image shows an autonomous vehicle with the technology. The vehicle is moving quickly and is being stopped by a building. The X-ray style vision lets the vehicle detect the van and put on the brakes to avoid a collision
Right, a pedestrian is “about to make an error in judgment” by walking into traffic. CP would allow a vehicle brake in time to avoid a collision.
Professor Nebot stated that this was done seconds before the driver’s local perception sensors. This could have been because the driver may have seen the same pedestrian around the corner. This would have given the driver extra time to react to the safety hazard.
In a real-life setting, CP would allow a moving vehicle to know that a pedestrian is about to walk out in front of traffic – perhaps because they’re too busy looking at their phone – and brake in time to stop a collision.
In this sense, the X-ray vision is an example showing how an autonomous car would be able to surpass the capabilities of a regular car driven by a human.
However, autonomous vehicle technology is still learning how to master many of the basics – including recognising dark-skinned faces in the dark.
Professor Paul Alexander is the chief technical officer at Cohda Wireless. He stated that the new technology has the potential to increase safety in situations with both autonomous and human-operated cars.
Safety remains a major concern for autonomous vehicles that have been subject to multiple trials around the world. Some self-driving cars were involved in human deaths
‘CP enables smart vehicles to break physical and practical limitations on onboard perception sensors, and embrace enhanced perception quality and robustness’ he stated.
“This could allow for a lower per-vehicle cost to facilitate large-scale deployment of CAV [connected and automated vehicles]Technology.
2021 was previously touted as the year fully automated vehicles would rollout on UK roads – but the technology is still in the trial phase.
Last year, Oxbotica, an Oxford-based autonomous vehicle software firm, launched a test fleet of six self-driving Ford Mondeos in the city.
The vehicles were each fitted with a dozen cameras, three Lidar sensors and two radar sensors, giving the fleet ‘level 4’ – the ability to handle almost all situations itself.
In the UK, a new lanekeeping technology that was approved under United Nations regulations was implemented in January.
This effectively means that vehicles can now be fitted with an Automated Lane Keeping System. Keeps the vehicle in its lane and controls its movements over extended periods of times without the driver being required to do anything.
The driver must be able and ready to take over driving control when the vehicle prompts. However, it would allow drivers the freedom to cruise on the motorway while texting or watching a film.
Potentially, car manufacturers would need to install shaking seats to alert drivers if they had to take over the vehicle.
The ALKS system is classified by the UN as Level 3 automation – the third of five steps towards fully-autonomous vehicles.
Six self-driving Ford Mondeos drove through Oxford at all hours to test the capabilities of driverless cars in a 2020 trial.
Safety remains a major concern for autonomous cars, which have been subjected to numerous trials worldwide.
Several self-driving cars have been involved in nasty accidents – in March 2018, for example, an autonomous Uber vehicle killed a female pedestrian crossing the street in Tempe, Arizona in the US.
According to reports, the Uber engineer was watching videos on her smartphone at the time.