By Lance Eliot, the AI Trends Insider
The new phonebook is here! The new phonebook is here!
That’s what people use to exclaim when the latest iteration of the local phonebook came out and was available for use. It was exciting since the venerated document contained the latest info about how to get ahold of someone. For those of you that perchance remember the antics of Steve Martin in the movie The Jerk, I’m sure you are already chuckling at the mention of a new phonebook and the accompanying delight that oftentimes arose due to the latest release of a thick tome that merely contained updated phone numbers and names.
It might seem like an oddball reaction nowadays, yet there was some validity to the keen interest in seeing the revisions and updates that arose in the latest incarnation of your local phonebook.
There is similar effervescent excitement going on for those that are deeply versed in the realm of self-driving cars, and it has to do with a kind-of newly iterated phonebook of sorts. Actually, in this case, the document is a cornerstone standard that many consider the Rosetta Stone for delineating the nature of self-driving capabilities, known as the SAE International standard and stuffily entitled (here’s a mouthful, get ready): “Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles.”
Those that are in the know refer to the standard by its official designation, namely J3016.
Within the high-tech field of self-driving cars, there is, of course, a breathless heralding of any brand-new updated release of this famous (or some bitterly say infamous) J3016 defining standard for ground-based autonomous vehicles.
Having been just posted recently, those within the self-driving car realm are undoubtedly poring through the document, seeking to find what types of changes and revisions have been enacted.
I’ll be happy to bring you up-to-speed.
Before any of you start to pull out your hair that perhaps there is an entirely new definition or taxonomy for self-driving cars, the bottom line is that this is a relatively tame set of revisions. The result is a tidier, more approachable version of the revered standard.
On the other hand, in case you don’t already know, the reason that I earlier referred to the document as the “famous or infamous” standard is that some within the self-driving car industry think the whole thing ought to be redone. Yes, start from scratch, they fervently say. There are a plethora of arguments about the number of levels of autonomy and how to best define the nature of self-driving or driverless technology.
You can easily engage in a heated debate on the J3016 that will cause your ears to wear out and your eyes to glaze over. Some insist it is a valuable keystone and thank goodness it was established and put into the world for all to use. Others decry it as either misguided or plainly out of touch.
Anyway, for those that are wanting a complete overhaul of J3016, you need to prepare yourself for the stony fact that this is not at all a do-over. I know that some pundits will lament that it is akin to merely moving around the chairs on the deck of the Titanic, but anyway, it is what it is. Save those bellicose arguments for another day.
Meanwhile, I’ll provide a smattering of highlights about the revisions made. I urge interested readers to peruse the new standard as they might so desire.
Before I get into the various changes, it would be useful to make sure we are all on the same page about what the self-driving car’s standard generally conveys. There are six levels identified that range from zero to five. Many of you have undoubtedly heard that some levels are considered true self-driving, and some levels that are not.
Allow me a moment to briefly indicate what that portends.
For my framework about AI autonomous cars, see the link here: https://aitrends.com/ai-insider/framework-ai-self-driving-driverless-cars-big-picture/
Why this is a moonshot effort, see my explanation here: https://aitrends.com/ai-insider/self-driving-car-mother-ai-projects-moonshot/
For more about the levels as a type of Richter scale, see my discussion here: https://aitrends.com/ai-insider/richter-scale-levels-self-driving-cars/
For the argument about bifurcating the levels, see my explanation here: https://aitrends.com/ai-insider/reframing-ai-levels-for-self-driving-cars-bifurcation-of-autonomy/
Understanding The Levels Of Self-Driving Cars
As a clarification, true self-driving cars are ones where the AI drives the car entirely on its own and there isn’t any human assistance during the driving task.
These driverless vehicles are considered Level 4 and Level 5, while a car that requires a human driver to co-share the driving effort is usually considered at Level 2 or Level 3. The cars that co-share the driving task are described as being semi-autonomous, and typically contain a variety of automated add-on’s that are referred to as ADAS (Advanced Driver-Assistance Systems).
There is not yet a true self-driving car at Level 5, which we don’t yet even know if this will be possible to achieve, and nor how long it will take to get there.
Meanwhile, the Level 4 efforts are gradually trying to get some traction by undergoing very narrow and selective public roadway trials, though there is controversy over whether this testing should be allowed per se (we are all life-or-death guinea pigs in an experiment taking place on our highways and byways, some contend).
Since semi-autonomous cars require a human driver, the adoption of those types of cars won’t be markedly different from driving conventional vehicles, so there’s not much new per se to cover about them on this topic (though, as you’ll see in a moment, the points next made are generally applicable).
For semi-autonomous cars, it is important that the public needs to be forewarned about a disturbing aspect that’s been arising lately, namely that despite those human drivers that keep posting videos of themselves falling asleep at the wheel of a Level 2 or Level 3 car, we all need to avoid being misled into believing that the driver can take away their attention from the driving task while driving a semi-autonomous car.
You are the responsible party for the driving actions of the vehicle, regardless of how much automation might be tossed into a Level 2 or Level 3.
For why remote piloting or operating of self-driving cars is generally eschewed, see my explanation here: https://aitrends.com/ai-insider/remote-piloting-is-a-self-driving-car-crutch/
To be wary of fake news about self-driving cars, see my tips here: https://aitrends.com/ai-insider/ai-fake-news-about-self-driving-cars/
The ethical implications of AI driving systems are significant, see my indication here: http://aitrends.com/selfdrivingcars/ethically-ambiguous-self-driving-cars/
Be aware of the pitfalls of normalization of deviance when it comes to self-driving cars, here’s my call to arms: https://aitrends.com/ai-insider/normalization-of-deviance-endangers-ai-self-driving-cars/
Self-Driving Cars And The Updated Standard
For Level 4 and Level 5 true self-driving vehicles, there won’t be a human driver involved in the driving task. All occupants will be passengers; the AI is doing the driving.
One aspect to immediately discuss entails the fact that the AI involved in today’s AI driving systems is not sentient. In other words, the AI is altogether a collective of computer-based programming and algorithms, and most assuredly not able to reason in the same manner that humans can.
Why this added emphasis about the AI not being sentient?
Because I want to underscore that when discussing the role of the AI driving system, I am not ascribing human qualities to the AI. Please be aware that there is an ongoing and dangerous tendency these days to anthropomorphize AI. In essence, people are assigning human-like sentience to today’s AI, despite the undeniable and inarguable fact that no such AI exists as yet.
Now that we’ve laid the stage appropriately, time to dive into the myriad aspects that come into play regarding the latest revisions of the lionized J3016.
To make sure we are all aligned, this discussion entails the officially labeled J3016 APR2021 revision, released at the tail-end of April 2021, and will be forever known as the J3016 2021-04.
For those of you that have a well-worn copy of the prior latest version, which came out in June 2018, you can toss that into the trash. There is a new king in town. If you don’t believe me, then maybe this quote from the J3016 2021-04 will convince you: “This edition cancels and supersedes the SAE J3016_201806.”
Clunk, that’s the sound of those old copies being placed into the recycling bin.
Moving on, the stated rationale for the standard has gotten a bit of an update. The key being that the opening rationale now provides pointed insights about why this April 2021 version exists and the prior June 2018 version is considered outdated.
Indeed, we can let the standard speak for itself in that it begins with this explanation about the overarching semblance of this revision: “This revision of SAE J3016 was undertaken in close cooperation between the SAE On-Road Automated Driving (ORAD) Committee and ISO TC204/WG14 through a Joint Working Group formed in 2018. This collaboration brought to bear the knowledge and expertise of global experts in driving automation technology and safety. Several new terms and definitions have been added and multiple corrections and clarifications have been made to address frequently misunderstood concepts and improve the utility of the document, especially for non-native English speakers.”
There is a bit of potential argumentation even for that presumably helpful introduction, but I won’t get into the weeds on that herein. Ask me after I’ve had a few beers.
There are plenty of backstories for just about every change that exists.
Here’s a newly added pronouncement at the start of the document that might seem entirely innocuous to the untrained eye: “Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. ISO and SAE International shall not be held responsible for identifying any or all such patent rights.”
Okay, no big deal, they added a cautionary notation about the nature of patents. You probably would have skipped right past that paragraph.
I’ll give you a glimpse into the cavalcade of backstories that are hidden behind the scenes and only insiders would be aware of.
As an expert witness in Intellectual Property (IP) rights cases involving autonomous vehicles and various AI technologies, I can tell you that there is a slew of patent fights that will inevitably engulf the entirety of the self-driving cars field. Nobody realizes this, as yet (I’ve mentioned this repeatedly in my columns as a wake-up call).
It will be an IP lawyer’s billable hour bonanza, for sure.
The skinny is that there are a boatload of patents that have been granted in this tech space. Many of the hardworking heads-down self-driving tech developers and engineers are (at this time) blissfully unaware of those patents. As such, there are tons of development and engineering efforts that are likely infringing on those patents, for which the firms making the self-driving cars are not yet cognizant.
Eventually, once self-driving cars start to be rolled out and become prevalent, they will become huge moneymakers. When they are making money, it will become worthwhile for the patent holders to dredge up their related patents and take those infringers to court (some refer to this disparagingly as being patent trolls). The odds are that much of this will be dealt with via out-of-court settlements, encompassing eyebrow-raising topnotch dollars being handed over, though occurring only once the specter of big court wins looms over the self-driving car makers and their fleet operators.
Given that context, the potential basis for the J3016 stating that it has nothing to do with those patents is to try and stay out of the morass of IP lawsuits and entanglements that will someday be a tidal wave. There are already some that believe the standard somehow entails or overlaps onto their patents, rightfully or wrongly, and thus the matter has already inched its way into view. The added narrative about the patents is presumably a get out of jail free card, put in place now and will be cited in the future by either side of those bruising patent battles to arise.
The main point here is that the subtlest of revisions and additions are more than what perhaps meets the eye. Entire and extensive complexities and intricacies are at the underpinnings of most of the changes.
The standard is akin to an iceberg whereby the only thing the naked eye is bound to see is the teensy-weensy tip of the iceberg. Be assured, there is a heap of fractious ice underneath.
Let’s take a quick look at another artful addition: “SAE Executive Standards Committee Rules provide that: This document is published to advance the state of technical and engineering sciences. The use of this document is entirely voluntary, and its applicability and suitability for any particular use, including any patent infringement arising therefrom, is the sole responsibility of the user.”
There is at least a twofer in that portion.
You can readily detect the patent topic. The other aspect is the notion of the standard being of a voluntary arrangement and that people can take it or leave it.
Due to space constraints, I can’t go into the entirety of the alluded to conflagration, but the gist is that some assert that the standard has been overplayed by some in industry and somehow become a type of “requirement” rather than something mere of an informative and industry helpful manner, i.e., being voluntary.
Enough said on that for now.
See if you can guess what this addition is about: “Any trade name used in this document is information given for the convenience of users and does not constitute an endorsement.”
Did you make your guess?
As you might have surmised, the concern about any type of promulgated and independent standard is that if it includes references to existing products or companies there is a chance that this might imply an endorsement for those so named. This would be a potential windfall for the named parties. It would also be a quite disconcerting slap in the face to other products or firms that believe they should be getting an equal footing.
Thus, it is best to excise such references as much as possible, averting the conundrum, or at least provide an escape clause that states any such references are not to be construed as an endorsement.
So far, none of those modest changes have much to do with the actual acts of self-driving. I’ll shift into that mode and give you a taste of the more technical nuances of the embodied revisions. One technical term that you’ll need to know is that DDT refers to the Dynamic Driving Task, which for herein simplicity’s sake you can blankly consider to be the act or task of driving a car.
See if you get the behind-the-scenes nature of this added line in the passage about what is considered inside of scope and outside of scope for the standard: “In addition, systems that inform, alert, or warn the driver about hazards in the driving environment are also outside the scope of this driving automation taxonomy, as they neither automate part or all of the DDT, nor change the driver’s role in performance of the DDT (see 8.13).”
It might be helpful if I give you the passage that just precedes this added line: “Active safety systems, such as electronic stability control (ESC) and automatic emergency braking (AEB), and certain types of driver assistance systems, such as lane-keeping assistance (LKA), are excluded from the scope of this driving automation taxonomy because they do not perform part or all of the DDT on a sustained basis, but rather provide momentary intervention during potentially hazardous situations. Due to the momentary nature of the actions of active safety systems, their intervention does not change or eliminate the role of the driver in performing part or all of the DDT, and thus are not considered to be driving automation, even though they perform automated functions.”
I realize that if you aren’t steeped in the nature of self-driving cars, those mentioned passages might seem like a strange foreign language, maybe a variant of Klingon.
In brief, there is an open question as to the boundaries associated with the automated and autonomous operation of a car. Imagine a person sitting at the steering wheel. They are responsible for the driving act. All of a sudden, someone outside the car yells out that there is a dog loose in the roadway. Hopefully, the driver hears the exhortation, mindfully considers it, and then steers the car to successfully avoid the beloved pooch.
I ask you, did the warning by the person outside the vehicle cause the vehicle to veer away from the canine?
Most would reasonably agree that the yelling didn’t change the actions of the car, and instead it was the driver that changed the actions of the car. The helpful shouting was useful to the driver, but in the end, it was the driver that was doing the driving. In that way of consideration, we could assert that the screaming was out-of-scope of the driving task. The driving task was still in the hands of the driver and the outside stimulus was nothing more than if the driver had seen the dog with their own eyes.
Just to let you know, not everyone would concur that the outside stimulus belongs outside of scope, and there is a rather complicated philosophical discussion that can ensue. In any case, for purposes of J3016, the revision now clearly states that any “systems that inform, alert, or warn the driver about hazards in the driving environment” are outside the scope.
I’m sure that you are entranced to now know that.
Let’s next consider an example of how there has been some significant technical rewording that has taken place.
The rewording in this example involves the definition of a commonly used phrase within the self-driving specialty, referring to the notion of achieving a Minimal Risk Condition (MRC).
Here is the prior definition of MRC: “A condition to which a user or an ADS may bring a vehicle after performing the DDT fallback in order to reduce the risk of a crash when a given trip cannot or should not be completed” (from the old section 3.17).
Here is the new definition of MRC: “A stable, stopped condition to which a user or an ADS may bring a vehicle after performing the DDT fallback in order to reduce the risk of a crash when a given trip cannot or should not be continued” (from the new section 3.16).
Though this might seem like a nonchalant or perhaps trivial difference, it actually is quite notable and proffers a substantive change that either can be interpreted in the context of the subsection as merely clarifying what was originally the overall intent about MRC or has perhaps given MRC a new meaning.
Studious self-driving devotees (or, some might prefer the cherished title of self-driving nerds), will be going toe-to-toe on that seemingly innocuous change.
Speaking of nerds or techies (again, meant in a most honored way), those of you that are staunchly familiar with the J3016 June 2018 standard will be somewhat taken aback at the aspect that the sections and subsections have been essentially renumbered (somewhat), thus the numbering that you memorized over these last several years will no longer do. For example, the portion on the MRC is an indication of that kind of change in numbering, previously being labeled as 3.17 and now it is 3.16.
But, hey, the good news is that this new version of the standard has finally earned itself a veritable table of contents, something sorely needed and well appreciated having been added.
Diehard fans of the June 2018 version will notice that the ADS-Dedicated Vehicle section 3.3 and the Conventional Vehicle 3.5 are now no longer in place. Meanwhile, you’ll want to check out an interesting new section 3.1.3 on Fleet Operations. And section 3.9 on Driving Automation System Feature Or Application has gotten a decidedly notable oil change and engine overhaul, as it were. Within the new incarnation of 3.9, now known as 3.7, there are these added subsections: 3.7.1 Maneuver-Based Feature, 3.7.2 Sub-Trip Feature, and 3.7.3 Full-Trip Feature. Make sure to give those depictions some close attention and focused reading.
I could go on and on, but I believe this gives you a taste of what the revised standard consists of.
For more details about ODDs, see my indication at this link here: https://www.aitrends.com/ai-insider/amalgamating-of-operational-design-domains-odds-for-ai-self-driving-cars/
On the topic of off-road self-driving cars, here’s my details elicitation: https://www.aitrends.com/ai-insider/off-roading-as-a-challenging-use-case-for-ai-autonomous-cars/
I’ve urged that there must be a Chief Safety Officer at self-driving car makers, here’s the scoop: https://www.aitrends.com/ai-insider/chief-safety-officers-needed-in-ai-the-case-of-ai-self-driving-cars/
Expect that lawsuits are going to gradually become a significant part of the self-driving car industry, see my explanatory details here: http://aitrends.com/selfdrivingcars/self-driving-car-lawsuits-bonanza-ahead/
If you are someone that is deeply dedicated to the advent of self-driving cars, you’ll want to get the J3016 2021-04 standard and have your red marker handy. It is well worth your time to pore through the revised standard and make sure that you know the ins and outs of what is currently the reigning king.
This will also require reorienting your prior predisposition of being able to quote verbatim from the June 2018 version. You don’t want to get caught by your colleagues or smarmy others that will stop you mid-statement and rightly point out that you are no longer with it. That you have fallen behind. That you are living in the past.
By and large, almost everyone other than those that are nose deep into the details about self-driving cars will not especially know and nor particularly care about these revisions. It will be as though the Jedi master has told those neophytes to walk on past, nothing to see here.
For those of you that do want more analysis about the revised standard, I’m planning on doing so, especially if this initial analysis strikes a chord and finds overall interest by readers.
Meanwhile, aficionados are going to immediately have a field day with this new phonebook, oops, this newly revised standard. This is the type of material that can be used to get a vigorous and enthralling discussion going at the drop of a hat. All I ask is that please don’t go to fisticuffs. Let your words be enough and no need to get into any rough and tumble about the latest version of the acclaimed standard.
Keep in mind, there is likely to be another iteration in the making, and thus another future version, though if it comes out in the same deliberative interval of time, you’ll need to mark the year 2024 on your schedule and be on the look for it then.
I know this much: I can’t wait.
Copyright 2021 Dr. Lance Eliot http://ai-selfdriving-cars.libsyn.com/website
Read MoreAI Trends