A look inside TomTom’s partnership with HELLA Aglaia for crowdsourced HD maps

The path to a safer, cleaner, autonomous future starts with HD maps. In 2019, TomTom and HELLA Aglaia started a collaboration that aimed to deliver accurate localization and crowdsourced updates that would improve TomTom HD Map quality even further.

The need for high-quality HD maps

As we move towards an autonomous future, industry analysts agree that HD maps are essential to making safe autonomous driving possible. The TomTom HD Maps are next-generation maps that deliver highly accurate, up-to-date and realistic representations of the road. HD Maps improve localization to centimeter-level accuracy and sensor perception, which leads to safer path planning by automated driving systems.

Our HD Maps are made and maintained with a multi-sourcing approach, terabytes of high-quality data coming from survey vehicles, as well as accurate crowdsourced car sensor data and other professional sources. All of these are necessary to create a rich and high-quality HD Map that you can trust.

A multi-source approach to HD mapmaking also minimizes the time between a road change taking place in real life and the moment that change is reflected on the digital map. Referred to as reality to map time (R2M), the shorter R2M is, the more accurate the HD map and the safer it is for AD systems to rely on.

In this article, we will look at the proof of concept developed together with HELLA Aglaia and some promising first results.

A successful partnership

For over two decades, HELLA Aglaia have been committed to finding smart solutions for complex problems and implementing those to further the automated mobility of the future. In 2019, HELLA Aglaia and TomTom partnered to enable robust camera-based localization and crowdsourced HD map updates, all with the goal to bring faster and more accurate updates to HD maps.

With HELLA Aglaia supplying their mono camera, we performed test drives near Berlin.

How it works

Let’s take a look at how HELLA Aglaia and TomTom technologies work together to facilitate real-time updates for HD maps.

For the proof of concept, HELLA Aglaia supplied their mono camera and TomTom brought the latest TomTom HD Maps to perform test drives near Berlin.

The layers of information included in the HD Map are:

• Lane information, such as lane borders, lane dividers centerline and lane type information.

• Local rules and restrictions such as speed limits and whether overtaking maneuvers are permitted or not.

• RoadDNA, TomTom’s sensor-agnostic localization suite, including landmark-based localization with traffic signs, used by HELLA Aglaia’s localization software.

All of these layers must be as accurate and detailed as possible. Using Hella Aglaia’s in-vehicle camera software, processed camera data can be shared to TomTom’s cloud, updating the TomTom HD map in real time.

Localization

For precise localization, the HELLA Aglaia camera first does perception using machine learning for the detection and classification of lane dividers and traffic signs. The next step is to correlate these lane and sign observations with the TomTom HD Map.

Roadagrams

Next, real-life detections in the form of Roadagrams are sent to the TomTom Cloud. Roadagrams are accurate crowdsourced map snippets from car sensors. These Roadagrams are aligned and aggregated, and then compared with the HD Map. Detected changes in the map – such as new traffic signs – are processed and added as new attributes to the HD Map.

Map delivery through AutoStream

Updated HD maps can be delivered to the vehicle via TomTom AutoStream. AutoStream is our proprietary HD Map delivery system where vehicles build a horizon for the road ahead by automatically streaming the latest map data.

One step closer to an autonomous future

TomTom’s collaboration with HELLA Aglaia has shown yet another step forward towards autonomous driving, where we have proven the concept of crowdsourced HD maps to substantiate our multi-source HD mapmaking process. Using crowdsourced input to minimize R2M time, this collaboration advances the reality of a safe, comfortable and autonomous future.