Autonomous car startup Argo AI is giving researchers free access to their HD maps – TechCrunch


Argo AI is releasing curated data along with high-definition maps for researchers for free, the latest auto industry company to give up some of the information it captured while developing and testing stand-alone cars.

The goal, says the Ford Motor Company, is to give academic researchers the ability to study the impact that HD maps have on perception and prediction, how to identify and track objects on the road, and predict where those objects will move in seconds. future. In short, Argo sees this as a way to encourage more research and hopefully advancements in the technology of autonomous vehicles.

Argo has tagged this data collection and maps of Argoverse, which is being released for free. Argo is not releasing everything he has. This is a curated data set, after all. Still, it's a lot big enough to give researchers something to investigate and model.

This "Argoverse" contains a selection of data, including two HD maps with center lane lines, traffic direction and ground height collected on roads in Pittsburgh and Miami.

For example, Argoverse also has a set of motion prediction data with 3D tracking annotation for 113 scenes and more than 300,000 vehicle trajectories, including unprotected left turns and track changes, and provides a reference to promote testing, teaching and learning, according to the website. There is also an API for connecting map data with sensor information.

Argo is not the first independent vehicle company to open its data or other tools. But the company says that this batch of data is unique because it includes HD maps, which are considered one of the critical components for standalone vehicles.

Much of the attention in the world of stand-alone vehicles lies in the "brain" of the car. But they also need maps to provide information that will help these vehicles, whether they are operating in a depot or on public roads, the safest and most efficient route possible.

For researchers, access to these types of maps can be used to develop forecasting and tracking methods.

Earlier this week, Cruise announced that it would share a data visualization tool that created called Webviz. The web-based application collects raw data from all sensors in a robot – a category that includes autonomous vehicles – and transforms this binary code into visual elements. cruise The tool allows users to configure different panel layouts, each displaying information such as text logs, 2D graphics, and 3D representations of the AV environment.

And last year, Aptiv launched nuScenes, a large-scale dataset of a set of autonomous vehicle sensors.


Source link