The Society of Automotive Engineers just updated their guidance for testing automated driving systems. The report, J3018, costs $81 to view and summary seems more concerned with what the report does not cover as what it does.

“This document provides safety-relevant guidance for on-road testing of vehicles being operated by prototype conditional, high, and full (Levels 3 to 5) ADS, as defined by SAE J3016. It does not include guidance for evaluating the performance of post-production ADS-equipped vehicles. Moreover, this guidance only addresses testing of ADS-operated vehicles as overseen by in-vehicle fallback test drivers (IFTD).

These guidelines do not address:

– Remote driving, including remote fallback test driving of prototype ADS-operated test vehicles in driverless operation. (Note: The term “remote fallback test driver” is included as a defined term herein and is intended to be addressed in a future iteration of this document. However, at this time, too little is published or known about this type of testing to provide even preliminary guidance.)

– Testing of driver support features (i.e., Levels 1 and 2), which rely on a human driver to perform part of the dynamic driving task (DDT) and to supervise the driving automation feature’s performance in real time. (Refer to SAE J3016.)

– Closed-course testing.

– Simulation testing (except for training purposes).

– Component-level testing.”

I have not purchased the report, so I can’t speak toward its contents. I can say that the SAE Autonomy Levels, although controversial, have been tremendously helpful in establishing a vocabulary for the industry.

Lots of automotive groups are grasping toward any sort of guidance around testing and certification. Getting the industry to converge seems like a key milestone in deploying production-level self-driving cars to the public.

The SAE might gain more by suggesting testing guidelines in the public domain, rather than behind a paywall.

Indeed, the world’s largest job search engine, reports that Aptiv is posting far and away the most self-driving car job openings. Aptiv spun out from Delphi, a Tier 1 auto supplier, a few years ago. They also acquired nuTonomy, which at the time of acquisition had logged more self-driving car miles than all but a handful of competitors.

Following Aptiv on the list is NVIDIA, and then SAIC Innovation Center (the Chinese auto supplier, not the US government contractor of the same acronym).

It would be interesting to get an absolute number of jobs from this report, rather than a percentage. When I search Indeed for jobs at SAIC Innovation Center, I only see a handful of listings. But maybe the number has waxed and waned over time.

Bosch and Daimler, which are partnering hand and glove on autonomous technology, come in at 4 and 5.

Meanwhile, Cruise Automation, which is has plastered hiring billboards up and down highway 101 in Silicon Valley, only rings at number 6, followed by GM.

The rankings don’t really match my intuition of who is hiring and how much, but presumably Indeed has data that I don’t.

The report also lists the top geographies for these jobs.

Silicon Valley — Technically this covers two places on the list: #1 San Jose and #4 San Francisco. A combination of Stanford, UC-Berkeley, Google, and a lot of venture capital and software engineers has made this the new center of robotics.

Pittsburgh — Carnegie Mellon University has spawned a bunch of corporate robotics teams, notably Uber ATG, Argo AI and Aurora.

Detroit — The center of the automotive industry, dating back to Henry Ford.

Portland — I assume this is mainly on the list due to Daimler Trucks?

Boston — MIT has not spun out quite as many robotics efforts as CMU, but nuTonomy counts for a lot, and there are other groups, too.

Baidu’s Apollo program has announced a self-driving car test in Changsha, Hunan, China, that is open to the general public. The program features 45 self-driving L4 Hongqi EV vehicles, the result of a joint partnership with FAW Group.

It’s always a little hard to know exactly what’s happening in China, because the non-Chinese press struggles with both access and language. The articles I’ve read use phrases like, “debuted”, “are being released”, and “kicked off”. Sounds like the program is already underway?

There is a human safety driver onboard, of course, and the geofence is limited to about 30 miles, although Baidu intends to expand.

Self-driving car deployments are trickling out, albeit not quite as fast as many of us had hoped. Opening up programs to the general public, as opposed to a limited and pre-screened group, is a big step forward.

I’m excited that Baidu is doing this. And I would love to see a video or writeup of what it’s like.

My Udacity colleagues Leah Wiedenmann and Vienna Harvey compiled a list of female leaders in the automotive industry. Their goal is to help more conferences invite female speakers.

I imagine this list will be helpful to journalists and researchers, as well.

https://docs.google.com/document/d/1DVxzLiwSfrRsoHQmYoU0Db7_WDy4ZUOwsWZBmjkW7IA/edit?usp=sharing

Automotive News has a cool article about different advances in automotive radar for autonomous vehicles. The article makes the point that radar has been a little bit like the forgotten sensor:

“Perhaps radar is even underappreciated. Venture capital has flowed into lidar and camera-based solutions for automated vehicles; radar has been viewed as a commodity.”

That seems right to me.

The article highlights three companies working on different approaches to more advanced radar for self-driving cars. The work from Bosch to create radar-based high-definition maps seems particularly interesting.

“By coupling these two inputs [radar and GPS], Bosch’s system can take that real-time data and compare it to its base map, match patterns between the two, and determine its location with centimeter-level accuracy.”

Bosch calls this approach, “radar road signature” and posits that it can provide centimeter-level accuracy while using half as much data as a camera-based map.

Bosch is highlighting their work with TomTom to build radar-based HD maps. They divide these maps into three layers:

• Localization: calculating where the car is in the world
• Planning: deciding which actions are available to the car
• Dynamic: predicting what other actors in the environment will do

This is exciting work because high-definition (HD) maps are usually the domain of lidar. Lidar point clouds are used to generate maps against which a vehicle can compare later sensor readings.

Some work has gone into attempts to build such maps with camera data. Visual SLAM is one example of this. By comparison, relatively little work has gone into building HD maps from radar data. That makes Bosch’s endeavor novel and exciting.

Bosch is positioning this as a fleet-based mapping system, with map data generated by ordinary consumer cars, not necessarily specialized mapping vehicles. It’s hard to know how realistic that really is, but it would play to Bosch’s strength of scale.

“One million vehicles will keep the high-resolution map up to date.”

As the world’s largest automotive supplier, Bosch has a unique ability to pump a success into the automotive market.