Waymo Open Dataset To Enable Free Sharing Of Self-Driving Data For Research

Knowledge Hub
Automotive Performance Engineering

Powered by

Waymo Open Dataset To Enable Free Sharing Of Self-Driving Data For Research

Waymo Open Dataset Free Sharing Self-Driving Data Research Autonomous Driving Future Development

Company makes its autonomous driving dataset open to use for future development

Waymo, self-driving vehicle innovator, has released the Waymo Open Dataset, which comprises of high-quality multimodal sensor dataset for autonomous driving, to the research community at large. This dataset is available free to researchers at waymo.com/open, comprising of high-resolution sensor data collected by Waymo self-driving vehicles. The dataset covers a wide variety of environments, such as dense urban centres, suburban landscapes, as well as data collected during various times of the day and in different weather conditions.

Webinar on Accelerating design decisions on hybrid/electric vehicle architectures with predictive simulation

The Waymo self-driving vehicles have collected over 10 million autonomous miles in 25 cities, the company noted. It said this rich and diverse set of real world experiences has helped its engineers and researchers develop Waymo’s self-driving technology and innovative models and algorithms. Waymo believes this is one of the largest, richest, and most diverse self-driving datasets ever released for research.

Click here for video on Designing Autonomous Vehicle for Series Production, Webinar Highlights

Waymo Open Dataset contains data from 1,000 driving segments, with each segment having captured 20 seconds of continuous driving, corresponding to 200,000 frames at 10 Hz per sensor. Such continuous footage gives researchers the opportunity to develop models to track and predict the behaviour of other road users, the company explained. Each segment contains sensor data from five high-resolution Waymo LIDAR and five front-and-side-facing cameras, providing a high-resolution, 360° view.

Waymo Open Dataset (2).jpg

The open dataset also includes LIDAR frames and images with vehicles, pedestrians, cyclists, and signage being carefully labelled, capturing a total of 12 million 3D labels and 1.2 million 2D labels. The company has also been working on 3D perception models that fuse data from multiple cameras and LIDAR.

Whitepaper on Integrating Hardware/Software Development for Autonomous Vehicles

Waymo said that when it comes to research in machine learning, having access to data can turn an idea into a real innovation. This data has the potential to help researchers make advances in 2D and 3D perception, and progress on areas such as domain adaptation, scene understanding and behaviour prediction, it added. The company further said that it hopes that the research community will generate more exciting directions with its data that will not only help to make self-driving vehicles more capable, but also impact other related fields and applications.