DeepMap Deep Dive: Year-End Insights from Analyst Brad Templeton

By | December 29, 2020

To cap off a rollercoaster of a year, we invited Brad Templeton to present a year-end Deep Dive to DeepMap employees. Brad discussed mapping technology for semi-autonomous and autonomous vehicles (AVs), and provided an analysis of significant events in the AV space in 2020.

Detailed Maps = Advanced Knowledge
Brad spoke about the importance of highly-detailed maps, such as those from DeepMap, for semi-autonomous and autonomous driving. This type of map is designed to be read by machines, rather than humans. It provides advanced knowledge about the road to inform driving decisions — in essence, allowing the car to see around the corner.

Brad pointed that when we hear the word “map,” we often think about the what of maps — the images on a screen or paper — rather than the why of maps — which is to give us information about the road that we would not be able to reliably figure out in the moment, while driving at a high speed.

Even when not looking at a map per se, human drivers get knowledge of the road through “map information” in the form of lanes marked with left turn arrows, dotted lane lines, and signs showing that a winding road or blind corner is ahead, for example.

Road engineers are continually painting and updating “map information” onto highways and city streets. As human drivers, we rely on this knowledge, especially when driving fast. AVs will also benefit — in terms of safety and reliability — from highly-detailed map information from multiple sensors.

Brad said: “Think of it this way: without a map or with a wrong map you may get to “Level X” of safety in a semi-autonomous or autonomous vehicle. A highly-detailed map will get you to “Level X Plus. While all cars must handle roads that have changed, a car with detailed maps drives at Level X Plus almost all the time, and on rare occasions drops down to Level X. A car without a comprehensive map is always at the lower level.”

2020: Winter to Spring
Brad described the autonomous vehicle space in 2020 as a “winter to spring story.” The year started on a somber note, with Starsky Trucking shutting down and Drive.ai being sold as an acquihire. Some automakers pulled back on full self-driving (Level 4) projects to focus on advanced driver-assistance systems (ADAS).

By mid-year, we started to see a “thaw,” with significant AV announcements energizing the market and creating optimism about the upcoming new year. Brad counts the following events in his list of 2020’s most notable AV highlights:

Zoox autonomous vehicle in San Francisco. Source: Zoox
  • The Zoox Reveal
    Zoox, which recently sold to Amazon, had been secretive about its vehicle design but finally unveiled it. It has face-to-face seats and sliding doors, as well as extra touches like sensors with no blind spots.
  • “No-Driver” Milestones
    Cruise, Yandex, AutoX, and Nuro reached the goal of putting cars on the road without a human driver behind the wheel.
  • Waymo’s Safety Report and Robotaxi
    Waymo published a report on safety incidents from over 6 million miles of driving in suburban Phoenix. This is comparable to over 8 human lifetimes without being at fault in an accident, a superhuman level of performance. They also released the world’s first real robotaxi service in Chandler, AZ.
  • Excitement over LiDAR
    The LiDAR space is exploding, with SPAC-based public offerings by Velodyne and Luminar — with huge valuations. Expect more in 2021.
  • Tesla’s FSD Beta
    Much attention was paid to Tesla’s limited release of its “full self driving” (FSD) beta. On YouTube you will see hundreds of videos taken by Tesla fans (showing the good and the bad). Tesla is leading the way in ADAS progress.

Now, onto 2021! Thank you, Brad, for a thought-provoking talk to help us welcome in the new year!

* * *

About Brad Templeton: Brad Templeton is founding faculty for Computing & Networks at Singularity University, and Chairman Emeritus and futurist of the Electronic Frontier Foundation (EFF), the leading cyberspace civil rights foundation. He is on the board of the Foresight Institute He also advised Google’s team developing self-driving cars, and writes about such cars at robocars.com and https://www.forbes.com/sites/bradtempleton/. He also advises Starship on delivery robots and Quanergy in the LIDAR space, plus companies in the micromobilty (scooter), minimobility and e-VTOL (flying car) areas. He founded ClariNet Communications Corp (the world’s first “dot-com” company.) He also created rec.humor.funny, and its web site, www.netfunny.com, the world’s longest running blog.

* * *

About DeepMap: DeepMap is accelerating safe autonomy by providing the world’s best autonomous mapping and localization solutions. DeepMap delivers the technology necessary for self-driving vehicles to navigate in a complex and unpredictable environment. The company addresses three important elements: precise high-definition (HD) mapping, ultra-accurate real-time localization, and the server-side infrastructure to support massive global scaling. DeepMap was founded in 2016 and is headquartered in Palo Alto, Calif., with offices in Beijing and Guangzhou, China. Investors include Andreessen Horowitz, Accel, GSR Ventures, Generation, Goldman Sachs, NVIDIA, and Robert Bosch Venture Capital. For more information, see www.deepmap.ai.


DeepMap Deep Dive: Year-End Insights from Analyst Brad Templeton was originally published in DeepMap Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.

Read the original article

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.