Sign In

Communications of the ACM

ACM TechNews

Lyft Releases Self-Driving Vehicle Data Set, Launches $30,000 Challenge

View as: Print Mobile App Share:
predicted locations of cars based on the Prediction Dataset, illustration

The prediction solution is trained on over two million samples of location data of cars and other traffic agents contained within the dataset.

Credit: Lyft

Lyft has released a Prediction Dataset, which contains the logs of movements of cars, pedestrians, and other obstacles encountered by its 23 autonomous vehicles in Palo Alto, Calif. The company also plans to launch a $30,000 challenge in which participants will work to predict the motion of traffic agents.

The data set includes logs of more than 1,000 hours of traffic agent movement, 170,000 scenes (about 25 second each), 16,000 miles of data from public roads, 15,000 semantic map annotations, and the underlying HD semantic map of the area. It contains the information needed to develop prediction models that could enable autonomous vehicles to choose safe trajectories in given scenarios.

Testing and validation sets will be released as part of the competition, which will begin in August on Google's Kaggle platform.

From VentureBeat
View Full Article


Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA




No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account