Autonomous Vehicle Engineering - July 2021 - 18

ADAS
"
There is no such thing as Level 2-Plus or Level 3- or 4-Plus; we tried to
make that very clear. There are no in-betweens!
"
J3016 has not been without controversy since
its debut in 2014, some would say because of
a lack of clarity.
You're right, it has been fairly controversial from the
start. Everybody has an opinion when it comes to
language! The point I try to make, and that people have
criticized, is J3016 is a convention. It could be other
than it is, but at least it's comprehensive and internally
coherent. I no longer refrain from pointing out, when
people argue about J3016, that we couldn't even be
having these arguments without this framework and
language at this point!
Revisions to SAE Levels of Driving Automation standard include clearer
distinctions between Level 3 and Level 4 automation.
was a Cybersecurity engineering standard [ISO/SAE
21434, in 2020]. Around the same time, we also started
a collaboration with ISO, in this case TC204, the ITS
side of ISO, again to produce a joint standard of J3016.
I believe this was driven by ISO's desire to harmonize
with SAE on these Levels of Driving Automation so
the industry can finally get on the same page and be referencing
the same terms, definitions and framework. The
point being that when we're doing important documents
like regulations, standards or guidance documents, it's
important that people use precise terms and definitions.
Were there some sticking points or misalignment
between the North American vehicle
development community and others - globally
- that prompted the J3016 update?
The short answer is yes. There was a fair amount of
North American jargon in the previous version of
J3016 which is simply a function of Americans being
the primary authors of that document. We did take
care of a lot of that. In addition, we wanted to provide
guidance and advice on things like logical groupings
of terms - in some places we have 'superterms' and
'subterms' that relate to the superterm. We've done a
better job with that.
It must be a challenge to put a 'live' document
like J3016 together at a time of such dynamism
and competition in the technology - in sensors,
processors, electrification. In many ways, you're
trying to hit a moving target.
The way we're handling this inside of SAE: J3016 is
sort of the 'Constitution,' if you will, of driving-automation
definitions. There are other documents
that have plenty of definitions - for example, J3131
that will be published shortly provides definitions
and taxonomy for [automated driving] architecture.
There is a taxonomy and definitions document
coming out of the SAE Verification and Validations
committee; there's J3164 coming out of the Maneuvers
and Behaviors committee that has new definitions.
There are a lot of expansions going on in terms and
definitions, to help nail down scenarios and situations.
So, we're trying to keep J3016 at its current high
'Constitution' level and allow the other SAE task forces
that are going into specific, deeper areas to formulate
their own definitions consistent with J3016. So
far, it seems to be working out fairly well. The nice
thing about this latest SAE-ISO version of J3016 (also
called ISO-PAS 22736), is it tells the world that we're
harmonized on the fundamentals. It's how we're going
to characterize this technology.
Perhaps the collaboration between the two
global standards groups will help mitigate
some of the early criticisms of J3016.
I hope so.
18 July 2021
AUTONOMOUS VEHICLE ENGINEERING
Continental
Autonomous Vehicle Engineering - July 2021

Table of Contents for the Digital Edition of Autonomous Vehicle Engineering - July 2021
