Some of the world’s top engineering minds are helping to design a fully autonomous transportation system for Saudi Arabia’s Neom megacity.
Fleets of self-driving electric vehicles will use advanced radar technology and communicate with each other and the city itself to help them one day transport goods and people without any need for human drivers.
The project is the brainchild of Dr Nahid Sidki, a Syrian-born engineer and pioneering roboticist who has a 30 year track record of delivering trail-blazing autonomous vehicle projects.
A former executive director of the Stanford Research Institute’s prestigious robotics centre, Dr Sidki is now the chief technology officer of the Research Products Development Company (RPDC), a Riyadh-based innovation centre and Taqnia subsidiary that is executing the mission of the of the Saudi Arabia Advanced Research Alliance (Sara), a network of research and development organisations from both the public and private sectors, which counts Aramco among its founding members.
Dr Sidki has assembled an international team of designers, technicians and materials scientists for the project, which aims to produce vehicles that are capable of level five autonomy - the highest level and a target that has so far eluded the world’s top engineers.
The transportation system Dr Sidki and his team are working on is designed to function without any human input at all, setting it apart from most of the self-driving cars in development around the world.
In order to achieve this, Dr Sidki’s team plan to implement new, high-tech sensor and networking technology.
Unlike the self-driving cars being built by automotive industry giants like BMW, Volvo, Nissan and Tesla, which rely heavily on lidar sensors to scan for obstacles, the Neom inspired vehicles will use radar sensors embedded and distributed around the bodywork with city-to-vehicle integration to achieve the highest levels of safety and reliability.
Dr Sidki told The National that the lidar systems currently being used would be unsuitable for terrain like the deserts of Saudi Arabia and could leave a self-driving car vulnerable to failure.
“Lidar is a very good sensor, but it has a lot of limitations,” he said. “The sensor has limited range and performs poorly in the rain, in fog, or in a sandstorm. Here in Saudi Arabia we have frequent sandstorms. If that sensor failed, the whole system would fail.”
Dr Sidki said that while companies like Tesla are pouring plenty of money into developing clever software, not enough resources are being devoted to improving the sensors required for navigation.
“Most of these companies are investing billions in the autonomy and the software, but not so much in advanced sensor development.”
Though the project is still in the early stages of a five-year development cycle, Dr Sidki’s team has already built a prototype miniaturised radar system that could be printed on a non-metallic surface like a car body.
In July of last year, Tesla’s Elon Musk told the World Artificial Intelligence Conference in Shanghai that he was “extremely confident” that the first fully autonomous vehicles were on the horizon.
But experts say significant hurdles to the development of the technology remain.
Daniel Faggella, the founder of AI research company Emerj, said the wide variety of conditions that any self-driving car is likely to encounter was a key hurdle for engineers hoping to build a system capable of full autonomy.
“Vehicles have to operate in the day time, night time, in snow, sleet, rain or hail,” he said.
“The diversity of not only the cars that have to move, but the objects, items, people and vehicles that are around them is so great and so vast that handling those edge cases is very challenging.”
A further hurdle that has so far held back the development of fully autonomous vehicles, he added, was the high levels of precision and reliability they need to be able to operate safely.
High levels of safety will be required for passengers and transport authorities to accept self-driving vehicles, he said.
To reach greater operating safety, the prototype being developed will also feature decentralized computing and use 5G to communicate with other vehicles and with the smart city itself.
RPDC, which was founded in 2015 as a subsidiary of Taqnia, (a PIF-owned company) is the national centre for technology development and commercialisation, brings together academic and industrial research centres as part of an effort to provide an end-to-end innovation solution and convert the Kingdom’s innovation into breakthrough technology.
The innovation hub has already supported the development of a robotic arm for inspecting undersea oil pipelines, a dangerous job usually carried out by human divers.