A new report by self-driving development organization Zenzic sets out the global standards for high-definition (HD) mapping when used in conjunction with autonomous driving.
Published in partnership with Ordnance Survey (OS), the UK’s national mapping agency, the ‘Geodata report – analysis and recommendations for self-driving vehicle testing’ also calls for the creation of common data standards that promote collaboration and improve confidence in mapping data for self-driving vehicles.
The report goes into depth on a number of issues including:
The level of detail required for self-driving vehicle mapping: OS and Zenzic have determined that self-driving vehicles will require maps with resolution better than 5cm to ensure vehicles can operate in complex environments. Maps will also need to include information on curbs, street-level features like lamp-posts, pedestrian crossings and road markings. Real-time updates to maps will also be crucial to let self-driving cars ‘see’ around corners for temporary objects in the road like skips or roadworks.
Why self-driving vehicles require a new generation of live maps: Self-driving cars use a range of sensors to ‘see’ the world around them. However, interpreting that information in real-time requires a lot of processing power. With high-definition maps that are updated in real-time, a self-driving vehicle is able to reference the position of other road users against what it already knows to be there. It also provides a backup in situations where its sensors are less effective. Adverse weather conditions like heavy rain or sun reflecting off a wet-road can make relying on sensor data alone difficult.
What standards will be necessary globally for self-driving mapping to be available and useful: Currently there is no single source of high-definition mapping data; each self-driving company is having to develop its own from the ground up. Ordnance Survey suggests a neutrally hosted platform for mapping data would increase the confidence in the data as it comes from multiple sources and would help different self-driving vehicles coexist on the same piece of road. For this to work, standards for how data is collected and shared will need to be implemented globally.
Daniel Ruiz, CEO, Zenzic, said, “The UK’s goal is to be able to benefit from self-driving vehicles on our roads at scale by 2030, a target that requires the development of technologies and tools that do not fully exist today. Our report with OS is another stake in the ground for the UK as a leader in the self-driving revolution and shows how the UK is building on its expertise in areas like mapping to drive the world forward.”
Simon Navin, head of innovation programs, OS, added, “The economic and societal benefits that can be achieved through the introduction of self-driving vehicles on UK roads should be significant. Through our work with Zenzic we are helping define the geospatial and mapping requirements that will accelerate the testing and adoption of self-driving technologies so that these benefits can be realized safely and efficiently.
“We believe that consistent, authoritative and trusted data provides a framework for safe operation, interoperability and open standards development. It will also enable innovative solutions from a wide range of providers who will bring new and exciting solutions to the UK mobility sector,” he said.