|A Range Rover Velar at the 2018 Geneva Auto Show. / Photo by: Alexander Migl via Wikimedia Commons|
Most car manufacturers focus their efforts on developing autonomous systems for on-road vehicles that would help passengers get around cities and roadways. While they are busy with that, British company Land Rover has set its eyes on off-road autonomous vehicles via the Cortex project.
Many car owners prefer Land Rover vehicles because of two main benefits: off-road aptitude and in-car comfort. However, There have yet to be models fitted with autonomous driving capacity. So, Land Rover decided to implement such technology to deliver a new driving experience.
The $5-million Cortex project will be dedicated to the creation of a self-driving system for off-road vehicles. The project consists of a research team at Land Rover and another team at the University of Birmingham in London. The researchers will employ the Myrtle AI program developed by a company in Cambridge. They chose this AI system to promote off-road predictability, which means no basic predictability of curbs, roads or lane markings.
As of the moment, the research team is working with the radar system for these vehicles. In on-road AVs, the radar system typically uses only 10 percent of its capacity to detect moving objects. But the Land Rover team desires to apply newer high-resolution radars from the market and find out the remaining 90 percent capacity of these systems.
In theory, it will allow the off-road AVs to see more than other vehicles around them with better resolution. However, the potential problem of the method is the processing of a large amount of data, which will require high-speed mobile data and a set of high-powered processors.
“The thrust of Cortex is how we can develop algorithms and smart processing techniques to retain as much of that information as possible, whilst making the big data issue containable,” explained Nigel Clarke, manager at Jaguar Land Rover, owner of the Land Rover brand.
The Cortex project has been limited to last for only 30 months. Although the team was given a short time, any significant progress during the project's course would be a positive result.