Stupidity, Damn Stupidity, and Lethal Features

Article By : Colin Barnden

GM's Ultra Cruise enables rich, lazy people to experiment with hands-free operation of 6,000-lb tanks. What a stupid idea.

At a recent investor event, General Motors (GM) announced “Ultra Cruise,” the successor to “Super Cruise,” its well-regarded Level 2 hands-free driving assistant. Suitable for use not just on divided highways but city streets too, Ultra Cruise enables rich, lazy people to ride in 6,000-lb tanks in close proximity to vulnerable road users (VRUs) without the onerous task of steering. What a stupid idea.

In a detailed post accompanying the announcement, GM described the functionality of the system as follows:

“Ultra Cruise will cover more than 2 million miles of roads at launch in the United States and Canada, with the capacity to grow up to more than 3.4 million miles. Customers will be able to travel truly hands free with Ultra Cruise across nearly every road including city streets, subdivision streets and paved rural roads, in addition to highways.”

Source: GM (Click on the image for a larger view.)

To “grow up” is precisely what GM management needs to do. This is such a stupid and reckless idea that I cannot believe a legacy automaker would even consider it, let alone announce it.

But what else should we realistically expect from the company that earlier this year attempted to sue long-term rival Ford for having the audacity to choose a product name featuring the industry-standard term “Cruise”? Get a grip, GM.

Why is GM following Tesla into the stupid phase of innovation? Money. Having said mostly crazy things for the last five years, on September 11 Tesla CEO Elon Musk wrote on Twitter:

“To be fair, investors are giving us significant credit for achieving self-driving, given that Tesla’s valuation/production is very high compared to other automakers.”

He’s right. Not that Tesla has achieved “self-driving,” but he has accurately observed that investors are giving Tesla significant credit for achieving it.

A cursory glance at the ranking of automakers by market capitalization reveals precisely why GM CEO Mary Barra seeks a sprinkling of innovation fairy dust. As of this writing, Tesla accounts for approximately one-third of total automaker market capitalization, but less than one percent of global light vehicle production. Which tells us a lot of Tesla investors are stupid too.

GM’s timing is woeful and Ultra Cruise’s future looks set to go up in flames faster than a Chevy Bolt. Why? Because the era of extremely accommodating federal policy for the deployment of experimental automated driving technology on public roads is drawing to a close.

Vulnerable road users

As Tesla’s experiments with Autopilot and “Full-Self Driving” show, not all innovation is safe innovation. This is why Tesla is now being investigated by the National Highway Traffic Safety Administration (NHTSA) and why the National Transportation Safety Board (NTSB) has had to conduct multiple crash investigations into Tesla and issue repeated safety recommendations.

According to data released by the National Safety Council (NSC) in March, 42,060 people are estimated to have died in motor vehicle crashes on U.S. roads in 2020. “It underscores the nation’s persistent failure to prioritize safety on the road,” the NSC observed.

In the release, the NSC identified ten life-saving measures that would improve road safety:

  • Equitable implementation of roadway safety laws, policies, procedures, infrastructure improvements is sorely needed. Communities of color are disproportionately impacted by motor vehicle crashes, roadway policies, lack of access to public transportation and poor infrastructure, among other critical elements that make a safe system.
  • Mandatory ignition interlocks for convicted drunk drivers, lowering state BAC levels to .05 and better education about the nature of impairment and when it begins.
  • Lowering – not raising – speed limits in accordance with roadway design, using a safe system approach.
  • Installation and use of automated enforcement to support safe speeds and adherence to traffic lights.
  • Laws banning all cell phone use – including hands-free – should extend to all drivers, not just teens. States with existing bans need to upgrade enforcement from secondary to primary.
  • Seat belt laws should be upgraded from secondary to primary enforcement and restraint laws should extend to every passenger in every seating position, in all kinds of vehicles.
  • All new drivers under 21 – not just those under 18 – should adhere to a three-tiered licensing system for novice drivers.
  • Automated Driver Assistance Systems (ADAS) that have life-saving potential should be standardized and accelerated into the fleet.
  • Motorcycle helmet laws should be passed or reinstated.
  • Communities and municipalities should adopt comprehensive programs for pedestrian and bicyclist safety.

Which sounds a lot like the NTSB’s 2021-2022 “Most Wanted List” of transportation safety improvements, shown below.

Source: NHTSA. (Click on the image for a larger view.)

Look carefully and you will observe that what is missing from both the NSC and NTSB list of safety improvements is Level 2 hands-free driving assistance on city streets. Why? Because road safety isn’t advanced by filling city streets with lazy drivers for whom the effort of steering a large luxury vehicle while close to cyclists and pedestrians is simply too burdensome.

Like Tesla, GM has failed to understand, or simply doesn’t care, that city streets are shared with vulnerable road users. In so doing, GM has likely just set itself onto an eventual collision course with the NTSB.

Hurricane Homendy

Led by new NTSB Chair Jennifer Homendy, we can already witness a dramatic shift in intervention by U.S.-based advocacy groups intent on redressing the imbalance between technological innovation and safety. Speaking in the cold open of the Autonocast, a podcast discussing the future of transportation, Homendy observed:

“I hear a lot about innovation in the news. I hear a lot about infrastructure investment and I’m not hearing a lot on safety. And we’re here to fill that void. I will be as vocal as I can be about it, because I do think there is this misperception of [Level 2] partial automation means it [the vehicle] can drive itself and I can just sit there.”

Translation: Homendy won’t watch quietly while automakers do stupid things to woo investors.

EE Times has written extensively about the practical problems of introducing driving automation on public roads, including mode confusiondriver roles, and driver responsibility. The problems are exacerbated by a mostly silent safety regulator which historically failed to provide guidance of the distinction between safe and unsafe innovation.

The common thread running through Homendy’s public comments is the need to protect VRUs. Viewed through the lens of safety rather than innovation, we can already see the new direction for safety advocacy in the U.S. which is away from soundbites and towards influencing policymaking.

Below are four changes to U.S. federal highway safety policy I believe could be implemented in the short-term:

  • No human supervision of Level 2 partial automation systems permitted in proximity to VRUs.

Level 2 operation enforced by federally designated operational design domain limits, restricting use to divided highways only. Tesla’s Autopilot and “Full-Self Driving” and GM’s Ultra Cruise are disabled for use on city streets via an over-the-air update.

  • Mandatory installation of intelligent speed adaptation and automatic emergency braking vulnerable road user (AEB-VRU) on all new highway vehicles with four or more wheels. AEB-VRU performance minimum equivalent to Euro NCAP test protocol.
  • Robust driver monitoring installed on all new highway vehicles with four or more wheels. Performance minimum might include 2-megapixel CMOS image sensor with 940nm IR optical path, operating at 60 frames per second tracking head pose and eye gaze. Signals derived from monitoring driver’s glance patterns to provide assessment of driver attention, distraction (both cognitive and visual), drowsiness, engagement, and impairment.
  • Operation and testing of Level 4 autonomous driving systems (ADS) banned on all public roads, with development limited to private test tracks and simulation. Federal guidance subsequently issued that precisely defines metrics for safe operation of Level 4 ADS technology on public roads.

While it is challenging to select the stupidest comment from Elon Musk over the last five years, a contender must be:

“If, in writing some article that’s negative, you effectively dissuade people from using an autonomous vehicle, you’re killing people.”

Which, according to Elon’s reasoning, clearly includes me. However I lose no sleep over his judgment, choosing NHTSA and NTSB  as the arbiters of safe innovation.

Tesla appears determined to discover that if a company develops an innovation strategy dependent on oversight in safety regulation, it cannot be surprised when those gaps in regulation are eventually closed. Stripped of Autopilot and “Full-Self Driving,” Tesla is reduced to a niche automaker with a cult following and a rapidly declining share of the electric vehicle market. Its current valuation looks extremely hard to justify based on those terms.

Why a respectable automaker such as GM would choose to adopt a similar strategy to Tesla is a mystery. Fortunately, there is still time to hit the brakes and steer onto a new heading, before GM’s headquarters at Renaissance Center in Detroit enter into the path of a hurricane.

This article was originally published on EE Times.

Colin Barnden is principal analyst at Semicast Research and has over 25 years of experience as an industry analyst. He is considered a world expert on market trends for automotive vision-based driver monitoring systems (DMS). He holds a B.Eng. (Hons) in Electrical & Electronic Engineering from Aston University in England and has covered the automotive electronics market since 1999.


Subscribe to Newsletter

Leave a comment