When you go to see a medical doctor, they sometimes use medical jargon that is hard to understand and leaves you feeling unsure and unable to engage in a lively discussion about your medical condition.
Most professions have their own set of lingo and specialized acronyms, providing a kind of shortcut way to express aspects in their specialty.
Anthropologists point out that besides using language as a shorthand, the vocabulary of specialists in a particular domain provides a cultural way of fitting in among fellow experts. Indeed, sometimes an expert will purposely brandish their lingo as a means of showboating and serve as a form of puffery.
If you try to dig into the lingo of a profession, the vocab frequently includes seemingly inscrutable alphabet soup-like acronyms.
It’s hard enough to grasp the meaning of one acronym, let alone make sense of them when the acronyms are stacked up one after another, turning a sentence into a scree of capital letters and appearing to be a strange foreign language.
To be clear, any such vocabulary does serve a quite useful purpose.
One vital purpose consists of being able to have coherent discourse, trying to ensure that if one person speaks of a rose, others all have an in-common understanding of what the word rose means. If one expert assumes that a rose means a rose, while a fellow colleague thinks a rose means an apple, the odds are that any discussion will breakdown right away as neither one agrees with the other.
The definitions of the vocabulary words, phrases, and acronyms in a domain are crucial since otherwise everyone is talking past each other and inexorable confusion will result.
Self-Driving Car Industry Jargon
The self-driving car industry is replete with numerous acronyms and specialized lingo.
Here’s a taste of some of the oft used acronyms: ADS, ODD, DDT, OEDR, OTA, V2X, ADAS, etc.
If you know what each of those means, kudos!
If you think maybe you know what each of those means, tip of the hat for your awareness.
If you don’t know what they mean or have heard or seen them but aren’t quite sure what they mean, welcome to the club since most people are in that same boat.
Now, some of the jargon associated with self-driving cars have relatively crisp definitions, but there is much lingo that is not well defined and remains elusive.
Unfortunately, the elusive portion tends to allow the media to use the lingo in ways that industry insiders find disquieting or downright upsetting (for the fake news aspects permeating the self-driving car field, see this link here).
Pile on top that even the well-defined vocab is repeatedly misrepresented by the media, including distorting the meaning or changing the meaning by using the lingo in ways that aren’t the intended use, and we’ve got a colossal communications mess on our hands.
A rose becomes an apple which becomes a gorilla.
How can the public be expected to realize what the true state of self-driving cars consists of, and where it is heading, when the words expressed are so sloppily bandied about?
How can regulators aim to provide suitable regulatory guidelines and rules if the meaning of the words and phrases of the industry are malleable and shapeless?
It’s a problem.
Part of the basis for the difficulty in assigning distinct and inarguable meaning to the lingo is that the field is generally new and evolving.
Unlike other professions that have had a long time to iron out their wording issues, the field of self-driving cars is relatively young and immature.
That’s not to say that there aren’t plentiful efforts underway to pin down the jargon.
I’ll be providing herein what I believe to be a consensus and informal style definition for many of the most commonly used acronyms and words, along with including some of the formalized versions of the definitions (especially citing what many consider the well-accepted definitions that are being promulgated by the Society of Automotive Engineers or SAE, and as especially reflected in the SAE J3016 standard).
There are those insiders that might have some heartburn with my informal definitions, and indeed there are insiders that already have heartburn over the formalized definitions too.
The hope here is to provide a semblance of what the vocab consists of, aiding all readers to come up-to-speed, and engage all in a dialogue about what the meaning is, could be, and ought to be.
That’s the caveat to keep in mind as you read the definitions.
A dictionary lists words in alphabetical order, which is handy since you can then quickly find a word by looking for it in its proper sequence.
Rather than using a dictionary sequencing approach, I’ll cover the jargon in a sequence that gradually builds upon each of the acronyms and phrases, doing so in a somewhat logical manner rather than a strictly alphabetic order. This will be more readily comprehensible and provide a gradual, progressive sense of order.
Trying to keep the list succinct, not all acronyms and not all words and phrases are being included. It is admittedly a cherry-picking of the most commonly used and relied upon ones.
Without further ado, let’s unpack the matter.
Untangling What We Shall Call A Self-Driving Car
The best place to start this jargon journey consists of the very cornerstone of how to say that you are referring to self-driving cars.
It’s an unsightly problem when everybody willy-nilly opts to use whatever phrases they wish when denoting an entire thing or field of inquiry (let alone the subordinated words and phrases).
Here are various ways that people refer to self-driving cars:
· Self-Driving Car
· Driverless Car
· Autonomous Car
· Robo-Car (also Robo-Taxi)
· Autonomous Vehicle (AV)
· Semi-Autonomous Vehicle
· Advanced Driver Assistance System (ADAS)
· Automated Driving System (ADS)
A wealth of riches in terminology, yet indubitably bankrupt since it is too loosey-goosey.
My preference and the preference by most that have considered the matter studiously is to normally use the phrase “self-driving cars” when referring to cars that are outfitted with automation that makes them able to drive without human intervention. It seems sensible to use because the “self-driving” portion implies that the car is able to drive itself, which is a handy and apt implication.
The phrase “driverless car” is somewhat problematic because it presumes that the word “driverless” is referring to the aspect that there isn’t a human driver, though the word “driverless” can also have a double meaning and imply that nothing at all is driving the car, which doesn’t make any sense since the automation is indeed doing the driving.
Presumably, the phrase “autonomous car” might be a better choice since it contains the word “autonomous” and therefore directly implies that the car is being driven by the automation, but it isn’t as catchy and seems flat on the tongue.
One of the most detested phrases is “robo-car” since it suggests that a robot is sitting in the driver’s seat and driving the car, though that’s not the path being primarily pursued (there are some efforts along those lines, though they are far and few between, see the link here).
The close cousin to robo-car is robo-taxi which is even worse since it not only has the same robot-implying element it also uses the word “taxi” and suggests that all self-driving cars will be used on a ridesharing basis (perhaps most will, but not necessarily all).
Researchers tend to use the phrase “autonomous vehicle” and its corresponding acronym of AV, though this is not seemingly as well-accepted by the general populace, plus the word “vehicle” is a broad term that encompasses cars, trucks, boats, submarines, drones, planes, and the like. Thus, using AV is not specific and makes it less useful if you are desirous of referring solely to self-driving cars.
On a related note, keep in mind that a car can be fully autonomous, or it can be partially autonomous (meaning that there is a human driver required and the automation supplements the human driver), which adds confusion to the matter and makes the wording additionally problematic.
The phrase “semi-autonomous car” helps somewhat to clarify that you are referring to a car that is less automated and thus presumably not a fully autonomous or true self-driving car.
Similarly, ADAS or Advanced Drivers Assistance System is indicative that the automation is intended to assist a human driver, rather than replace the human driver.
ADS or Automated Driving System is the formal terminology used by the SAE to refer to the driving system that covers levels 3 to 5 of the taxonomy of driving automation (I’ll explain the levels momentarily):
· ADS (Automated Driving System)
· Formal definition: “The hardware and software that are collectively capable of performing the entire dynamic driving task (DDT) on a sustained basis, regardless of whether it is limited to a specific operational design domain (ODD).”
Talking About Driving
This brings us then to the next logical juncture of wanting to define the levels of automation for driving, along with explaining what DDT and ODD are about.
Let’s start with DDT or the dynamic driving task, which boiled down means driving the car:
· DDT (Dynamic Driving Task)
· Formal definition: “All of the real-time operational and tactical functions required to operate a vehicle in on-road traffic, excluding the strategic functions such as trip scheduling and selection of destinations and waypoints.”
The ODD or operational design domain has to do with where and when the driving automation is capable of driving the car or its defined scope:
· ODD (Operational Design Domain)
· Formal definition: “Operating conditions under which a given driving automation system or feature thereof is specifically designed to function, including but not limited to, environmental, geographical, and time-of-day restrictions, and/or the requisite presence or absence of certain traffic or roadway characteristics.”
And the levels of driving automation are used to denote how capable the driving automation is, ranging from a score or rank of 0 to a topmost score of 5 (with the topmost score meaning the fullest capable).
· Level 0: No Driving Automation
· Formal definition: “The performance by the driver of the entire dynamic driving task, even when enhanced by active safety systems.”
· Level 1: Driver Assistance
· Formal definition: “The sustained and operational design domain-specific execution by a driving automation system of either the lateral or the longitudinal vehicle motion control sub-task of the dynamic driving task (but not both simultaneously) with the expectation that the driver performs the remainder of the DDT.”
· Level 2: Partial Driving Automation
· Formal definition: “The sustained and operational design domain-specific execution by a driving automation system of both the lateral and longitudinal vehicle motion control sub-tasks of the dynamic driving task with the expectation that the driver completes the object and event detection and response sub-task and supervises the driving automation system.”
· Level 3: Conditional Driving Automation
· Formal definition: “The sustained and operational design domain-specific performance by an automated driving system of the entire dynamic driving task with the expectation that the dynamic driving task fallback-ready user is receptive to an automated driving system issued requests to intervene, as well as to dynamic driving task performance relevant system failures in other vehicle systems, and will respond appropriately.”
· Level 4: High Driving Automation
· Formal definition: “The sustained and operational design domain-specific performance by an automated driving system of the entire dynamic driving task and dynamic driving task fallback without any expectation that a user will respond to a request to intervene.”
· Level 5: Full Driving Automation
· Formal definition: “The sustained and unconditional (i.e., not operational design domain-specific) performance by an automated driving system of the entire dynamic driving task and dynamic driving task fallback without any expectation that a user will respond to a request to intervene.”
In recap, there are lots of ways to refer to self-driving cars, of which the preference when mentioning a car that has automation allowing the car to drive itself is the phrase “self-driving cars” and then if the car requires a human driver it is referred to as a “semi-autonomous car” or one that is using ADAS (Advanced Driver Assistance System).
The act of driving is referred to as the DDT or dynamic driving task, and for establishing a scope of where the automation is capable of driving, such as say only with a specific city and only in daylight, it is the designated ODD or operational design domain.
And, we use the score of 0 to 5 to denote how capable the automation is, seeking to achieve the vaunted topmost score of 5, a fully autonomous car that can drive anywhere and anytime that a human could drive a car.
Lingo About Sensory Aspects
We can now move into the additional territory of phrases and acronyms that go beyond the aforementioned core.
To undertake the driving task, the automation uses various sensory devices such as cameras, radar, ultrasonic, thermal, LIDAR, and so on, which enable the AI to detect what’s going on around the car and act somewhat like the eyes and ears of a human driver.
The proper acronym for these automation eyes-and-ears, along with the act of using the collected sensory info to drive the car is OEDR:
· OEDR (Object and Event Detection and Response)
· Formal definition: “The sub-tasks of the dynamic driving task that include monitoring the driving environment (detecting, recognizing, and classifying objects and events and preparing to respond as needed) and executing an appropriate response to such objects and events (i.e., as needed to complete the dynamic driving task and/or dynamic driving task fallback).
A somewhat informal term is used to describe the act of bringing together the sensory collected data and trying to make it into a cohesive whole, usually referred to as sensor fusion:
· Informal definition: The act of bringing together or fusing the sensory data collected from a multitude of sensory devices, doing so during the driving task, along with reconciling conflicts and aligning elements to gain an overarching semblance of the driving scene.
A sensory device known as LIDAR has become quite popular for use on self-driving cars, though there is some controversy in that Tesla and Elon Musk eschew using LIDAR (for more about this contention, see the link here).
· Informal definition: A sensory device that is considered a mashup of light and radar, which emits a laser light beam and upon reflection attempts to detect the distance between itself and whatever object the light bounces off.
Let’s take a breath and mull over this added lingo.
Whenever you hear about or see a self-driving car, one aspect to be looking for consists of the type of sensors being used, along with how the data from those sensors are incorporated into the AI being able to drive the car.
The overall capability is the OEDR, while the sub-portion that deals with bringing together the sensory is the MDSF, and among the myriad of sensors could be the use of LIDAR.
Hopefully, that otherwise cryptic sentence now makes sense to you!
Lingo About Communications
A true self-driving car is intended to be able to work all on its own and not need to communicate with any other external system to figure out how to drive the car.
But, this doesn’t mean that you can’t supplement the on-board AI by allowing it to communicate with other external systems that might aid the driving task.
Here are some important terms:
· Connected Vehicle (CV)
· Informal definition: A car that has communications capabilities to connect with external systems.
· Informal definition: A means to allow a vehicle to communicate with a variety of external systems, including for example V2V, V2I, V2P.
· Informal definition: A means to allow a vehicle to communicate with another vehicle.
· Informal definition: A means to allow a vehicle to communicate with roadway infrastructure such as traffic lights, bridges, railroad crossings, etc.
· Informal definition: A means to allow a vehicle to electronically communicate with pedestrians.
· Informal definition: A means to allow a vehicle to communicate with a cloud-based system, enabling the on-board AI to receive updates and allowing the car to push sensory data and other info up to the cloud.
That’s quite a mouthful of acronyms.
Here’s a quick quiz for you.
See if this sentence makes sense to you:
A self-driving car used it’s OTA this morning and got its latest AI updates, and while on a later driving journey used V2V to coordinate with other self-driving cars to coordinate traffic activity, including having avoided a freeway wreck that was alerted via a V2I broadcast.
Did the use of OTA, V2V, and V2I make sense?
Now for your final exam, try to make sense of this:
The semi-autonomous car using ADAS was not as capable as the Level 4 ADS and lacked the AI needed to autonomously perform the DDT for a complex ODD, partially because the OEDR was insufficient and especially had limited MDSF, along with lacking LIDAR, though it was a CV and could potentially use the OTA to advance, and was equipped too with V2V though not yet able to handle V2P and nor V2I.
Did you get that?
There are many more acronyms and phrases pertinent to self-driving cars, but you’ve now earned your honorary badge of the self-driving cars 101 and ought to pat yourself on the back accordingly.