In a Tuesday morning blog post, Lyft revealed that it is harnessing data from its ride-sharing service to enhance the development and performance of its autonomous vehicles. Lyft’s Level 5 self-driving car program will deploy this data to create three-dimensional maps, make sense of the driving patterns humans use, and refine simulation tests.
The Level 5 self-driving car program will use data from various cars in the company’s Express Drive Program, which allows drivers to rent SUVs and cars instead of leasing them through conventional channels.
The Level 5 program will collect necessary data using forward-facing cameras installed on participating Express Drive vehicles. Autonomous cars in Palo Alto, California, as well as their accompanying safety vehicles, will also collect data for the project using similar hardware and methods.
In a blog post, Peter Ondruska (Lyft’s head of AV research), Luca Del Pero (Lyft Engineering Manager) and Hugo Grimmett (Lyft Project Manager) collectively noted, “While mapping operations teams can build 3D geometric maps for AVs, keeping them up-to-date and scaling their scope is a challenge. We’ve mapped thousands of miles thanks to the wide geographic coverage of the cars on our network. We’re able to continuously update our maps based on a constant stream of data that is immediately logged when a ride is completed.”
Lyft’s cameras will assist in compiling data for the type of situations that drivers deal with on a daily basis. By having a wealth of data on these situations, Lyft’s AI can become more useful and robust.
Lyft’s Level 5 self-driving car program also makes use of visual localization tech to keep tabs on the trajectory drivers use while they’re behind the wheel. This data helps Lyft’s AV keep the ideal lane position, which is not necessarily the center of the lane.