HD mapping ‘takes ADAS to the next level’, says TomTom
Experiencing a partially automated highway driving system today can be both rewarding and stressful at times. While in most scenarios these systems are highly capable at controlling acceleration, braking and steering within lanes, any sudden or unexpected changes to the driving environment can occasionally spook the system.
One of the factors behind this issue is that the car currently relies on visual information, with very little context about what it might encounter further on up the road; is the bend too tight, are the road markings obscured, is there ice on the road? TomTom believes that its high-definition (HD) map technology can make all the difference and avoid situations where the car deems it necessary to disengage a partially or fully automated feature, and hand control back to the driver without warning. While these systems emphasise that driver monitoring is required at all times, such instances can erode consumer confidence in automated capabilities.
TomTom recognises that HD maps are not a silver bullet to the issue, however, and as such have been designed to work in tandem with various other technologies as part of a complete autonomous system. “We don’t see HD maps as standalone products. They don’t exist in a vacuum and must be used as part of the automated system in a vehicle,” explains Tomaso Grossi, Senior Product Marketer at TomTom Automotive. “We believe in a closed loop system, whereby we produce the most accurate, robust and reliable maps in the Cloud, deliver these systems to the car, and then leverage multiple sources – such as different types of vehicle sensors – to ensure this map is up-to-date, safe and matches reality.”
One of the issues with in-car navigation systems today is the lead-time between a change to the road network – be it traffic, road works or a crash – and an update being validated and issued. For the driver, this means that the in-vehicle map frequently becomes out of date, and TomTom believes this needs to change.
HD map information helps to take ADAS to the next level; if you can build a system that leverages HD maps to make the car safer, more comfortable and more aware, then trust and driver adoption in these systems will increase
“The map is a living and ever-changing, ever-updating system,” says Grossi. “We’re exploring and experimenting with multiple sources to keep the map up to date, and using data in the Cloud to feed it back into vehicles, ideally in real time.” Initial targets are to be able to completely refresh and validate the map with lead times of just under a week, but real-time updates are the primary goal in some instances. For example, it is not strictly necessary for real-time updates to the entire map in one go; certain sections of road networks change far less frequently than others. Controlled access highways, for instance, are the least variable types of roads, with the addition of new lanes and lane markings occurring quite rarely. Conversely, city roads tend to see the highest variation of changes.
While the ideal goal is for real-time updates across the board, TomTom has found that – based on its expertise in navigation maps – a cycle time of a few days or even a week is sufficient for most roads. “There will be different lead times depending on the class of road and where these roads are located,” explains Grossi. “Depending on how often these roads change, we’ll need to have the map updated as quickly as possible. We already are taking this challenge head on with the typical maps we use in navigation systems today.”
TomTom has been collecting data through its RoadDNA technology, one of the core elements to HD mapping, which will allow autonomous vehicles to become aware of their environment, location and path on the road. Certain roadways in mainland Japan, the US and Western Europe have already been mapped out using this technology, and it is being leveraged by several OEMs’ autonomous prototype vehicles. As it stands, just over 380,000 kilometres of roadways across these regions have HD map coverage, up from 250,000 kilometres globally as of March 2017.
TomTom is not the only player developing advanced mapping technology, and a handful of start-ups are investigating how such systems can support autonomous vehicle systems. Localised HD map coverage may benefit those testing vehicles in specific areas, but in order to facilitate widespread deployment of autonomous vehicles, these maps need to go national, if not global in future.
“The ability to scale up and produce these maps in a highly efficient way whilst maintaining accuracy, robustness and reliability is key,” says Grossi. “If you look at the current ecosystem of autonomous driving, maps are becoming increasingly important. New players are entering this area, but many of these start-ups have an approach that is maybe not as scalable. When it comes to OEM demands, they might fall short in terms of scalability and robustness.”
There are three primary use cases for HD maps: accurate localisation, supporting the environmental perception of vehicle sensors, and the ability to plan the path of the vehicle as accurately and efficiently as possible. In short: to ensure the car knows where it is, where it is going, and any issues that may affect its course on the road.
The ability to scale up and produce these maps in a highly efficient way whilst maintaining accuracy, robustness and reliability is key
“It’s about making the system safer, more comfortable, and building trust into the system,” says Grossi. “Most people that have tried semi-autonomous systems on the market today, or who have had a ride in an autonomous vehicle, are very enthusiastic about the technology. But at the same time, they are also aware of the limitations.”
Regulations and technical developments aside, trust in autonomous driving systems is expected to be one of the main factors that influences consumer adoption of the technology. HD maps are believed to be a vital element to the overall system in ensuring that the car can not only anticipate potential dangers, but also leverage real-time understanding of road conditions out of sight. While autonomous vehicles of Level 4 and 5 capability on the scale developed by the Society of Automotive Engineers (SAE) are some way off, there will be many drivers today that desire an improved semi-autonomous experience. Grossi advises that TomTom’s HD maps and RoadDNA software can also assist the advanced driver assistance systems (ADAS) that underpin highway pilots today.
“Our technology extends the car’s view beyond the range of it sensors, and allows it to anticipate what might happen. The car may be travelling at a particular speed toward a certain curvature in the road, and it would be more comfortable and safer if the car slowed down in advance, rather than reacting at the last minute,” he concludes. “HD map information helps to take ADAS to the next level. If you can build a system that leverages HD maps to make the car safer, more comfortable and more aware, then trust and driver adoption in these systems will increase.”