Making Sense Of AI Autonomous Cars Going In Reverse And Reversible Computing 

3119
Writer Lance Eliot ponders whether AI self-driving cars will reconsider the act of driving in reverse, or whether the task will remain as it is today. (Credit: Getty Images) 

By Lance Eliot, the AI Trends Insider  

They say that time always marches forward.   

Does it?   

Right now, we assume that time can only flow in one direction, namely forward. Mankind has dreamt forever that it would be nifty if time could be reversed. There are a plethora of science fiction tales including books, short stories, movies, TV shows, poetry, and you name it that have sought to explore what could happen and what might be done if time could flow in reverse.   

Time reversibility is undoubtedly a fascinating topic.   

Shifting gears, you might be surprised to know that time reversibility has been given some strident attention in the realm of computers and computing machines, doing so in a field of study known as reversible computing. 

Most people have never heard of such a thing. Indeed, even those within the computer field are often taken aback to discover that research and attention are being put toward reversible computing. The usual reaction is in three stages. First, amazement that such a field of study exists. Second, a doubtfulness that it makes any sense to study the notion and that it might very well be an utter waste of time (a bit of a pun there). Third, curiosity about what exactly reversible computing is, and how it might be of use. 

Generally, there are two major ways to categorize reversible computing facets. One is a means of performing physical reversibility of computational activities, while the other has to do with undertaking a logical semblance of computational reversibility. These two categories tend to work hand-in-hand.   

A simple example might suffice to get you started on the topic of reversible computing.   

It seems that everyone nowadays knows that there are usually computer bits consisting of the binary values of 1 and 0. We could construct an electronic chip that would take as input a bit, either in the state of 1 or the state of 0, and produce as output the so-called opposite or inversion of the bit (i.e., if the input is a 1 then output a 0, while if the input is a 0 then output a 1). Those of you familiar with such matters would recognize this as the NOT operation (confusing perhaps that this is known as NOT, which might seem like not doing something, when in fact it will produce the outputs as mentioned herein). More commonly, this is referred to as an inverter.   

Could you receive the output of a NOT operation and make things go in reverse, returning to whatever you started with? Yes, it would certainly seem straightforward to do so. If the output was a 0, you know that the input must have been a 1, while if the output is 1 then you know that the input must have been a 0 (assuming of course that the operation worked flawlessly). Now, keep in mind that most electronic chips are not made to work in reverse. We have logically identified how to go in reverse, but the electronics might not be set up to allow a reverse physical operation to occur. As such, if we wanted this to happen in a real-world sense, we might need to adapt the electronics accordingly.   

A reversible electronic circuit, then, is a circuit that allows for bits to flow in either direction, forward or in reverse, through the circuitry. There is no longer a fixed indication of what is an input and what is output since either side of the circuitry could be considered available for input and equally available for output. 

Where this can be usefully employed involves a somewhat complex understanding of entropy in computing and computers. Generally, it is posited that a reversible computing or computer system could potentially consume much less energy than a conventional forwards-only computer. There would in theory be less heat dissipated for a reversible computing system. That’s important because computers are getting smaller and smaller, and limitations are being approached involving what to do about the heat production, for which the give-off of energy can adversely affect the electronics and otherwise undermine attempts to reduce the size of chips. It is hoped that the per Joule of energy that you can squeeze out of the computer hardware would be significantly enhanced via adopting reversible computing approaches.   

For those of you more interested in software than hardware, you might enjoy knowing that there is reversible computing with respect to aspects of software too. 

For example, most software engineers value the notion of being able to stepwise execute their computer programs when doing debugging. This is usually done in a forward motion, proceeding with each line of code to the next in sequence or the intended next line to be performed, moving from the start to the finish. Some debugging tools will allow you to also work in reverse, such that you can essentially go back to a prior line of code, retracting what has just happened, and thus invoke your code to effectively perform in a reverse direction (this is not as easy a thing to do as it might seem on the surface). 

If you stop to think about it, we are seemingly conditioned to always be thinking about going forward, and less so about going in reverse. We think of computers for example as going in a forward motion, performing a series of steps in a progression from start to finish. Rarely do you think about going from the end to the start. 

There’s another place that we tend to be primarily focused on going forward, entailing the driving of a car. 

It is a pretty good bet that you spend most of your time driving in a forward direction. Sure, you do use the reverse capability of the car, such as when backing out of a parking spot or trying to back down your driveway. The preponderance of your time though at the wheel is likely going forward and only a tiny fraction of the time involved in reverse driving. 

Some people are quite rusty at driving in reverse. You see them moving an inch at a time when backing up. They aren’t sure whether to look over their shoulder or whether to look in the rearview mirror. Their heads twist back and forth, trying to figure out what is behind them. Luckily, technology has progressed that many cars now have a back-up camera built-in, aiding the reverse driving chore. Besides the camera, there are at times sensors that scan the area behind the car as you are driving in reverse, attempting to alert you if there are any objects detected. Nonetheless, a lot of older cars still exist on our roadways and they generally lack the back-up camera and allied technologies. 

Here’s an additional twist for your day: Will AI-based self-driving cars reconsider the act of driving in reverse, or will the task of reverse driving remain as it is today? 

Let’s unpack the matter and see. 

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 that 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 a Level 4 and Level 5, while a car that requires a human driver to co-share the driving effort is usually considered at a 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 say).   

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: https://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 Going In Reverse  

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.   

Will the AI be able to drive the self-driving car in reverse?   

Yes, if the AI has been set up to do so, and for which you can reasonably expect that nearly all automakers and self-driving tech firms are putting such capabilities into place.   

There are some important caveats to keep in mind. 

First, by-and-large, the mainstay of the sensors on a self-driving car are oriented toward going forwards, not going in reverse. There is usually not as much sensory capability installed at going in reverse as there is in going forwards, which somewhat makes sense since presumably, a self-driving car is predominantly to be used going forwards versus going backward.   

Second, this means that even if the AI has been programmed or otherwise established to drive in reverse, it is doing so in a less than optimal way than it can drive going forwards. Presumably, the slow speeds usually involved in driving in reverse are sufficiently low enough that the amount of sensory gear will be good enough for reverse oriented driving. 

But, this is potentially a legal liability issue that might ultimately rear its ugly head, as it were. If a self-driving car, while going in reverse, runs into say a child or harms or kills someone, you can be sure that lawyers are going to be asking pointed questions about how the reverse driving capabilities were designed and implemented. A case could be made that any shortchanges on how the reverse driving was devised are tantamount to insufficiencies that contributed to whatever regrettable incident might have occurred. 

You might be tempted to assert that human drivers are relatively poor at driving in reverse, therefore if the AI is similarly insufficient at driving in reverse then we are no worse off than with human drivers. I doubt that logic will prevail. The public is likely to have higher expectations about the AI driving system, and especially so when going in reverse. In fact, people oftentimes assume that the AI to-date is safer at driving in reverse than humans are, which, debatably might be the case in comparison to rather newbies at driving, but not necessarily better than say human drivers that have honed their reverse driving skills.   

Another consideration to contemplate about cars is that the design and construction of most conventional cars are entirely oriented toward forward driving. You cannot drive in reverse in any prolonged manner and nor at the same heightened speeds and fluidity as you can when driving forwards (in most cars). This is not simply due to the human driver, but also due to how the reverse driving mechanisms are devised. 

Of course, it seems entirely sensible that the car is structured for forward driving, including that there is a front windshield and nothing comparable on the back (the back window is not usually the same as a windshield). Our seats for drivers are oriented toward the front of the car, and the driving controls are intended to be used while facing forward. 

Here’s a mind-bender for you.   

For true self-driving cars, the driving controls do not need to be available in the car per se, since the AI is doing the driving. Furthermore, the seats inside a self-driving car do not need to be facing in a fixed manner forward. The typical interior design for a self-driving car consists of seats that swivel, allowing the passengers to face in any desired direction. Plus, the seats are oftentimes designed to recline, allowing you to catch a snooze while the AI is doing the driving.   

The AI doesn’t especially care that there is a so-called forward and a so-called backward (going in reverse). It can be crafted to drive in either direction.   

As such, some question why we would use a conventional car design for the use in creating self-driving cars. Sure, it is easier to do things that way, since those car designs already exist. But, perhaps we ought to be reconsidering that a conventional car is oriented towards a forward driving role, and instead allow for an equal capability of self-driving cars to go in reverse as it does when proceeding forwards.   

Akin to the earlier point about electronic chips that might allow for reversibility, recall that there isn’t any designated side that has exclusively the inputs or exclusively the outputs, thus, perhaps our self-driving cars should be made to go in either direction too, doing so to an equal capacity.   

Some pointedly ask why should we continue to limit self-driving cars to drive in the ways that have been structured to accommodate human drivers? 

Some readers might realize that I’ve discussed this topic in prior articles, including pointing out that Zoox has consistently had a core goal of attaining self-driving cars that can equally proceed forward and “in reverse” (a quite sensible proposition, plus a rarity among competitors in terms of opting to avidly pursue this ambidextrous kind of capability).   

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: https://aitrends.com/selfdrivingcars/self-driving-car-lawsuits-bonanza-ahead/ 

Conclusion  

There is more to the debate about self-driving cars going forwards versus going in reverse.   

For example, if this suggests that the sensory devices need to be doubled up to allow for reverse driving, perhaps the added cost is unwarranted (on the other hand, given the potential liability exposure, this could be considered a must in that light).   

Also, some worry that since our driving world is shaped around human drivers driving forward, perhaps it is unsafe to have self-driving cars routinely aiming to drive in reverse, since our roads, parking lots, and even our mindset as pedestrians is not prepared for seeing en masse driving in reverse. 

So, which is it, does driving in reverse for self-driving cars seem revolutionary or evolutionary?   

This is an open question. 

Meanwhile, please make sure to look carefully when driving in reverse, and perhaps also keep your eyes open for self-driving cars coming down your neighborhood streets while in reverse.   

Don’t worry, it won’t be an optical illusion, it’s the real thing. 

 Copyright 2021 Dr. Lance Eliot  This content is originally posted on AI Trends. 

[Ed. Note: For reader’s interested in Dr. Eliot’s ongoing business analyses about the advent of self-driving cars, see his online Forbes column: https://forbes.com/sites/lanceeliot/] 

http://ai-selfdriving-cars.libsyn.com/website