When a self-driving vehicle is operating, it is gathering information about the world around it. Cameras and LiDAR help it identify vehicles, pedestrians, signs and, hopefully, anything else that might be out in or near the streets. Radar helps the vehicle keep track of how fast things are moving around it.
“Without all this data, self-driving cars wouldn’t even be able to leave a parking lot,” says Tony Lockwood, Autonomous Vehicle Manager, Virtual Driver Systems, Ford Motor Company. He observes that these vehicles need to process a constant stream of information to “safely navigate” their surroundings.
Leaving aside for the moment giant unresolved privacy issues and how companies can sell your personal information their benefit not yours, data is needed to help engineers and researchers create software that can teach self-driving vehicles how to analyze their environments. This software used to be called the driver.
Ford is now releasing a wide-ranging self-driving vehicle data-set to the academic and research community. “There’s no better way of promoting research and development than ensuring the academic community has the data it needs to create effective self-driving vehicle algorithms,” says Lockwood. It is also a way of potentially mitigating the financially ruinous corporate investments needed to make autonomous vehicles viable and affordable.
As part of this package, Ford is releasing data from multiple self-driving research vehicles collected for one year – separate from the work d is doing with Argo AI to develop a production-ready self-driving system. This data-set includes not only LiDAR and camera sensor data, GPS, and trajectory information, but also unique elements such as multi-vehicle data and 3D point cloud and ground reflectivity maps. A plug-in is also available that can easily visualize the data, which is offered in the ROS format.
Ford claims that there are a number of reasons why these data points are noteworthy to researchers.
- Since this data-set spans an entire year, it includes seasonal variations and varied environments throughout Metro Detroit. It has data from sunny, cloudy, and snowy days,
- And freeways, tunnels, residential complexes and neighborhoods, airports, and dense urban areas.
- Add construction zones and pedestrians, and researchers have access to diverse scenarios that self-driving vehicles will find themselves in.
Key here is quantity. Some datasets only offer data from a single vehicle, but sensor information from two vehicles can help researchers explore new scenarios, especially when the two encounter each other at different points along their respective routes.
- Right now, one vehicle has limited “vision” in terms of what it can see; you will note in our visualizations that some parts are not colored in, which is because the vehicle’s sensors could not penetrate those areas.
- But with multiple vehicles in the same general area, it is feasible one would detect things the others simply cannot, potentially opening up new routes for multi-vehicle communication, localization, perception, and path planning.
This data set has more horsepower, so to speak, including high-resolution time-stamped data from Ford test vehicles’ four LiDAR and seven cameras. It is claimed that this can help researchers explore solving perception problems that have been the bane of AVs and can result in fatalities.
Precise localization and actual usage data allow researchers to see exactly how accurate their algorithms are, giving them a baseline for performance they can measure against in their own research.
In addition to releasing 3D point cloud maps from our LiDAR, Ford is also “giving” the research community access to high-resolution 3D ground plane reflectivity maps. Together, these maps give researchers a comprehensive understanding of what our self-driving vehicles “see” in the world around them.
The whole point of this effort is to not only improve the way self-driving vehicles navigate their environment and interact with personal cars, pedestrians, and other self-driving vehicles, but also to support the next generation of engineers. Offering researchers, a comprehensive package of information will enable them to create advanced simulations based on real data — and we are excited to see how this will all be used.
Available through Ford collaboration with the Amazon open data program (do you trust Amazon, which won’t even protect it’s workers right now, to protect you?), one can find out more about our self-driving data package by visiting avdata.ford.com. The first set of data logs is available, and Ford will continue updating the site until all the logs are uploaded.