{"id":118799,"date":"2020-06-17T16:04:42","date_gmt":"2020-06-17T16:04:42","guid":{"rendered":"https:\/\/blogs.nvidia.com\/?p=45843"},"modified":"2020-06-17T16:04:42","modified_gmt":"2020-06-17T16:04:42","slug":"checking-the-rearview-mirror-nvidia-drive-labs-looks-back-at-year-of-self-driving-software-development","status":"publish","type":"post","link":"http:\/\/webdomino.net\/index.php\/nvidia\/checking-the-rearview-mirror-nvidia-drive-labs-looks-back-at-year-of-self-driving-software-development\/","title":{"rendered":"Checking the Rearview Mirror: NVIDIA DRIVE Labs Looks Back at Year of Self-Driving Software Development"},"content":{"rendered":"

The NVIDIA DRIVE Labs video series provides an inside look at how self-driving software is developed. One year and 20 episodes<\/a> later, it\u2019s clear there\u2019s nearly endless ground to cover.<\/p>\n

The series dives into topics ranging from 360-degree perception<\/a> to panoptic segmentation<\/a>, and even predicting the future<\/a>. Autonomous vehicles are one of the great computing challenges of our time, and we\u2019re approaching software development one building block at a time.<\/p>\n

DRIVE Labs is meant to inform and educate. Whether you\u2019re just beginning to learn about this transformative technology or have been working on it for a decade, the series is a window into what we at NVIDIA view as the most important development challenges and how we\u2019re approaching them for safer, more efficient transportation.<\/p>\n

Here\u2019s a brief look at what we\u2019ve covered this past year, and how we\u2019re planning for the road ahead.<\/p>\n

A Cross-Section of Perception Networks<\/b><\/h2>\n

Before a vehicle plans a path and executes a driving decision, it must be able to see and understand the entire environment around the vehicle.<\/p>\n

DRIVE Labs has detailed a variety of the deep neural networks responsible for vehicle perception. Our approach relies on redundant and diverse DNNs \u2014 our models cover a variety of capabilities, like detecting intersections<\/a>, detecting traffic lights and traffic signs<\/a> and understanding intersection structure<\/a>. They\u2019re also capable of multiple tasks, like spotting parking spaces<\/a> or detecting whether sensors are obstructed<\/a>.<\/p>\n