By Lance Eliot, the AI Trends Insider
I was eagerly awaiting making a right turn on a red light at a busy intersection that led onto the always hectic Pacific Coast Highway (PCH). Dutifully stopping behind the crosswalk line, I could not quite see the prevailing traffic that was coming over a hilly portion of PCH that fed down through this particular area. The layout meant that you could not be fully sure whether the lanes were free to drive into and at any moment a raging car might seemingly leap over the hill and barrel into you if you prematurely had opted to make the treacherous right turn.
Being in a hurry, I decided that I would just go for it and gunned my engine to swiftly propel my car through the right turn and into the lane of traffic that seemed to be empty and available. Unfortunately, in the split second that I had hit the gas, sure enough a car appeared suddenly at the crest of the hill and I could see that it was intent on using the lane that I was about to occupy. Some drivers in my shoes might proceed anyway and force the other driver to either hit their own brakes or swerve to another lane to avoid me, the lane intruder.
I didn’t want to chance the possibility of having the nearing car rearend my car, plus I would obviously have been in the wrong for my actions. I urgently moved my foot from the accelerator over to the brake and jammed down hard, forcing my car to come to a heart stopping halt. Yes, I was protruding somewhat into the lane, but it was not enough to cause the other car to undertake a diversion or hit their own brakes. They could sweep past my now half-in half-out car and proceed without skipping a beat.
The story might end there, and I suppose might have been unworthy perhaps of telling if that’s all that happened, except for one little and quite important fact. There was a car that was behind me, also waiting at the red light, and the driver had been apparently hopeful of making the right turn as I did so. This car had inched its way up to my rear bumper and was earnestly urging me to make the right turn swiftly. I’m sure you’ve had this kind of eager beaver behind you. They breath down your neck and get irritated and upset if you don’t make a highly expedient right turn on red.
Well, this other driver had assumed that since I was gunning my engine and leaping forward, they might as well do the same. I guess we were going to be like two sprinters that heard the starting gun go off and it was then a race to get underway. When I opted to hit my brakes, doing so suddenly, the driver behind me was caught by surprise.
Allow me a moment to carp about that driver. In my view, if they are going to follow behind me with nary any space allotted between the two of us, it is incumbent upon that driver to be fully ready to react in whatever way the driver ahead of them acts. It is a solemn golden rule of follow-the-leader in car driving. Anyone that is going to play the follow-the-leader game and do so without a safety cushion, they need to mirror the actions of the leader and do so with the same quickness of any efforts by the leader. Anyone that can’t do that should not be participating in follow-the-leader. Just wanted to get that off my chest.
In any case, the car behind me almost rammed into my car. Thankfully, it was almost and not an actual hit. The driver swerved to avoid hitting the back of my car and jammed against the curb where the right turn was. I could see in my rearview mirror that the driver was upset – at me, and was apparently cursing and making unmentionable finger movements expressing their displeasure at my actions.
I usually try to calculate the moves of any cars behind me in whatever driving actions I take, and likely normally would have realized the other driver was within an inch of my car and not therefore have rocketed forward, but in this case I was in a hurry and assumed the other driver would be savvy enough to realize the potential for a rapid braking due to the PCH hilly road shape. Too bad for them that they miscalculated. It didn’t look like there was any actual damage to their car and so I proceed to now continue the right turn and make my way to my looming appointment.
I did have in the back of mind a concern that this driver might get themselves worked up into a road rage. You know how that can happen in these kinds of circumstances. The driver was mad at me. They were in a hurry. I fully expected the driver to now also make the right turn, and then attempt to catch-up with me, coming alongside of my car to make more reprehensible gestures at me. Or, perhaps even try to run me off the road. You never know what crazy things people might do.
For my article about road rage, see: https://www.aitrends.com/selfdrivingcars/road-rage-and-ai-self-driving-cars/
For the foibles of human drivers, see my article: https://www.aitrends.com/selfdrivingcars/ten-human-driving-foibles-self-driving-car-deep-learning-counter-tactics/
When you think about the situation, I probably accelerated for just a fraction of a second or so. My car appeared to leap forward like a leopard. The total distance covered was just a handful of feet. Luckily my brakes were darned good and when I jammed my foot onto the brake pedal the car came to an amazingly crisp stop. You could almost say that the whole thing happened in the blink of an eye.
I’m not sure that any of the other nearby drivers noticed what happened. The car that was barreling down PCH that I had wanted to avoid hitting was probably just happy that my car was not in their way. Other cars at the intersection that were waiting to go were probably looking at the red light and expecting it to soon turn green. All in all, I’d bet that only me and the car behind me were aware of the dance that the two of us undertook.
I’ve told you this story to bring up something that is quite important and yet often overlooked as an aspect of human driving behavior, namely the use of micro-movements when driving.
A micro-movement is considered a somewhat subtle driving action that exists somewhere in the gray area between doing an overt and obvious driving maneuver and doing essentially no noticeable driving maneuvering at all.
Some Drivers Oblivious to Micro-Movements
Some human drivers are oblivious to the nature of micro-movements. These drivers are blind to noticing micro-movements and can’t readily tell you what a micro-movement looks like. Their attention to the driving task is based on wholescale movements and maneuvers, ones that are stand out like a brass band and are so obvious that you’d truly need to be blind to miss them. Anything less than the outright maneuver of a car is totally lost on them.
I remember one colleague at work that used to drive several of us out to lunch on Fridays. He drove his car like it was a Sherman tank. He was master of the roadway, and no one dared to get in his way. This vividly became evident on one occasion involving him and a car that appeared to be driven by someone lost and unfamiliar with the local streets.
The driver ahead of us was driving relatively slowly in terms of being just a tad below the posted speed limit. The car had several times started toward the curb as though the driver was thinking of stopping or parking, and yet the driver continued ahead. It seemed apparent to me and everyone else in the car that my colleague was driving that this other driver was trying to figure out where they were and wanted to maybe park at a specific address.
My clueless colleague did not notice the antics of the other car. Driving his car like a take-no-prisoners tank commander, he nearly rolled over the other car when it tried to make a few of those potential darts to the curb. If it had been me doing our lunch journey driving, I would have given the other car some added space to allow them to do these swaying kinds of motions. Plus, I would have swung into the left lane and gone around the other car, hoping to then avoid the situation entirely of dealing with the other car.
After we made it to the lunch, I asked my colleague if he had noticed the car that had been in front of us. What do you mean, he asked, what was to be noticed? When I told him about the micro-movements of the other driver, my colleague insisted that I was making it up. It was a gag, he insisted. Fortunately, my fellow lunch time eaters all chimed in about that car. They had all noticed it too. Of course, this only inspired my colleague to claim that it was a conspiracy and we were all ganging up on him.
For my article about conspiracies, see: https://www.aitrends.com/selfdrivingcars/conspiracy-theories-about-ai-self-driving-cars/
For the need to use defensive driving techniques, see: https://www.aitrends.com/selfdrivingcars/art-defensive-driving-key-self-driving-car-success/
You don’t necessarily need to be in a car to see the micro-movements of other cars. I ride my bike quite a bit and anyone that is hopeful of surviving the riding of a bike in city traffic is likely to be an expert on the micro-movement evidenced by cars and those wacky human drivers.
Just last week, I was in the bike lane, minding my own business, and I had a car that zoomed along and passed me, doing so with a flourish. The car traffic up ahead was stopping to deal with a roadway construction effort, and I lamented in my mind that this “jerk” was going too fast toward it. In my book, the driver ought to be slowing down and getting ready to figure out how to safety navigate around the street repair efforts.
Due to the bunching up of the car traffic, I caught up with the traffic since the bike lane was unimpeded. It was one of those rare cases whereby you can go faster along than the cars can. The cars were jockeying for position to find a means to navigate around the street crew and meanwhile I was rocketing ahead on my bike. Though I relish a moment of going faster than cars, I also know that doing so is tantamount to poking at a slumbering beast that can awaken at any moment and attack you.
Remember the car that had zoomed past me? It became the slumbering beast. The driver was anxious about being snagged in the snarled traffic. I could see that the car was nudging somewhat into the bike lane. At first, it was merely a slight intrusion. I had a hunch that the driver was testing the waters about whether or not he could come directly into the bike lane, doing so to use it as a means to escape the stunted traffic and get ahead of the other “sheep” that were stuck in the pile of cars.
I moved to the rightmost position of the bike lane and was kissing the curb. This was a defensive preparatory move by me. I also had my hands squarely on the brake handles of my bike, ready to use my brakes if needed. I observed the tires of the car that was inching into the bike lane and could see that the driver was becoming more decisive about angling toward the bike lane.
Sure enough, he then maneuvered completely into the bike lane, which was not only dangerous for me, but there were other bike riders in it too. The madcap driver completely upended all of the bike riders, not by the car actually hitting the riders, and instead by the bike riders all swerving this way or that way to avoid the intruding car, like bowling pins at a bowling alley. Have you ever seen those televised bike races in France and seen those moments when the bike riders appear to fall and collide all at once, it was about the same here. Sacre Bleu!
Sometimes a micro-movement can be signaled by something other than the car itself. For example, a human driver puts their arm out the driver’s side window to wave you around them, or maybe they point at something up ahead to forewarn you that there’s a big pothole in the road. Hand motions though can be confusing and at times not meant to be a car movement indication.
I watched the other day as one driver put his hand and arm fully out the window and seemed to be waving the appendage up and down, rather frantically, which I wasn’t sure what this might portend. Maybe the earth had opened a giant sinkhole and the road ahead was merely a huge ditch? When I opted to drive around the car, figuring I’d need to be on my toes and contend with whatever I might encounter, I could hear loud music being played on the speakers in the car — I surmised that the hand and arm motions were just his way of keeping a beat to the music.
A driver’s head can be another form of a clue about driving behavior. I’m sure you had times when you’ve been driving and could see the head of the driver in the car ahead of you. The driver perhaps turns their head to the right, and you are able to predict that the driver is going to try and either swing into the lane to the right or start to make a right turn. Another head signal involves a head nod, often used when at a four-way stop and the other driver wants to showcase that either you should go or they want to proceed.
For my article about the head nod problem in AI self-driving cars, see: https://www.aitrends.com/selfdrivingcars/head-nod-problem-ai-self-driving-cars/
For why humans are going to be pranking AI self-driving cars, see my article: https://www.aitrends.com/selfdrivingcars/pranking-of-ai-self-driving-cars/
For the dangerous practice of people jumping out of moving cars, see: https://www.aitrends.com/selfdrivingcars/shiggy-challenge-and-dangers-of-an-in-motion-ai-self-driving-car/
For aspects of how we use conspicuity when driving, see my article: https://www.aitrends.com/selfdrivingcars/conspicuity-self-driving-cars-overlooked-crucial-capability/
Let’s focus on the micro-movements of cars that are evidenced by the actual car and set aside for now the kinds of micro-movements that humans themselves might undertake with their own bodies.
Here are some the key ways that micro-movements can appear:
- Angling of the wheels of the car
- Leftward leaning of the car
- Rightward leaning of the car
- Stuttered motion of the car (start/stop)
- Car lurches (sudden motion)
- Riding of the brakes
These aforementioned car movement indications are considered micro-movements when they are done in a subtle manner. It’s a kind of a tease. Just a little bit of showing of the leg, as it were.
Some might liken this to a “tell” in a sport like basketball. When playing basketball competitively, you might give a movement of your head that seems to say you are going to make a run to the basket, and the other player covering you can potentially read the tell and get positioned to block you. This can all happen in an instant. It can happen too before the player with the ball has actually moved their feet toward the basket and taken any other overt actions.
Your Tell Gives You Away
If you like to play poker, you certainly know about the tell or clues that other players often inadvertently show. When I was in college, I played a Saturday poker game with some of my buddies. One member of the tight knit group was well-known for trying to bluff his way on hands where he had nothing of use in his cards. It was the classic saying about the notion that weak means strong, and strong means weak, namely that whenever he tried to make it seem like he had all aces, it really meant he had junk. This was his tell.
Another player had a tell that involved pushing his glasses up his nose and toward his eyes. We all eventually figured out that this was his tell. It meant that he was lying about whatever he might be saying. None of us revealed the tell to him, and we kept it each to ourselves, relishing it. After months of this, one of the pack opted to tell him about his tell.
At first, he was angry that we had noticed the tell. He was also angry that we had not revealed it to him, though he sheepishly realized that he probably would not have told any of us about our tells. It dawned on him that if we had become conditioned to his tell, he could use that conditioning to his advantage. For a few weeks, he would purposely push his glasses on his nose, doing so to make us think he was lying, even though he might be telling the truth. We got caught like a hungry fish on a shiny lure. He made some good money by exploiting his own tell.
One of the members of the group had been oblivious to the tell. When it was finally revealed to everyone at one of the poker games, this particular member was shocked. What tell? He had never noticed that our friend was often pushing his glasses up further on his nose.
This highlights my earlier point about micro-movements of cars. Some people notice them, some do not. And it also highlights another very important point. You cannot for sure bet on a micro-movement since there is no guarantee it will ultimately graduate into becoming a full-on movement. Not all tells blossom. You need to assess each one, along with the situation at-hand, and ascertain whether the micro-movement is telling you something of substance or not.
Let’s also make clear what isn’t a micro-movement. If someone completely cranks the wheels over to make a full turn, I don’t think we can really consider that to be a micro-movement. Only if the driver had just slightly angled the wheels, perhaps doing so to get ready for a turn, or maybe they were almost subconsciously starting the turn, those are situations of a micro-movement.
One of the ways to study micro-movements often involves observing teenage novice drivers. There are some newbie drivers that stretch out every driving move they make. When they are told to make a turn at the corner up ahead, they start the turn a quarter mile beforehand. When they see a stop sign, even if its at a distance of a football field length away, the novice driver often starts to tap those brakes, getting the car slowed down, slowly, really slowly.
Those tend to be exaggerated micro-movements. Most seasoned drivers are apt to make much more subtle micro-movements. Those weathered drivers have become ingrained in how they drive a car. They aren’t necessarily aware that they are using micro-movements.
Similar to my story about my poker playing friend that ended-up exploiting his tell, there are some savvy drivers that like to use their micro-movements in an exploitive way. These drivers often assume that other drivers will notice the micro-movement and then back-off or otherwise give way to the driver that is using the micro-movements.
This happens a lot in parking lots. When cars are driving around and around trying to find a parking spot, it can become a gamesmanship effort of who suggests they found an open spot first. You’ve likely seen two cars that came open an open spot, and the cars are facing each other head-to-head. Which one will get the spot? If one driver angles their wheels, it can be a kind of assertion that they have tagged the spot and are going to aggressively make their way into the spot.
This can at times create more problems than it solves.
Suppose you have two cars facing each other, trying to vie for a parking spot, and each makes a wheel angling micro-movement. Is this a tie game? Will one driver take Umbridge that the other driver appears to be claiming the spot?
We also need to consider the clueless driver that doesn’t even notice the other car is angling its wheels.
Or, you can also have the driver that sees the micro-movement and opts to ignore it, acting as though they didn’t notice it, so they can appear to be blissfully ignorant as they take the spot (and later claim they didn’t realize the other car was trying to go for it). That’s a sly move.
Besides teenage novice drivers showcasing micro-movements, there’s another kind of driver that can do likewise, namely a drunk or DUI (Driving Under the Influence) driver. When someone is just partially drunk, they often exhibit numerous micro-movements. They do so because they are either unsure of what driving path they are going to take, or they are lacking in their mental facilities and are erratically coping with the car controls.
If you are a seasoned driver that watches for drunk drivers, I’m sure you’ve seen many such micro-movements. That being said, once a drunk or DUI driver has gotten fully engulfed by their mind-altering influence, their car directing movements are bound to become the opposite of micro-movements. The driver make overt and extremely overstated movements. Any subtlety about their driving is now completely gone.
For more about drunk driving, see my article: https://www.aitrends.com/selfdrivingcars/dui-drunk-driving-self-driving-cars-prevention-cure/
For tit-for-tat driving behaviors, see my article: https://www.aitrends.com/selfdrivingcars/tit-for-tat-and-ai-self-driving-cars/
For my article about rocket man styles of driving, see: https://www.aitrends.com/selfdrivingcars/rocket-man-human-drivers-ai-self-driving-cars-outrunning/
For the role of greed in driver behavior, see: https://www.aitrends.com/selfdrivingcars/selfishness-self-driving-cars-ai-greed-good/
What does this have to do with AI self-driving cars?
At the Cybernetic AI Self-Driving Car Institute, we are developing AI software for self-driving cars. One aspect that is not yet getting much attention involves the micro-movements of driver behavior, which some consider an edge or corner case problem.
Allow me to elaborate.
I’d like to first clarify and introduce the notion that there are varying levels of AI self-driving cars. The topmost level is considered Level 5. A Level 5 self-driving car is one that is being driven by the AI and there is no human driver involved. For the design of Level 5 self-driving cars, the auto makers are even removing the gas pedal, brake pedal, and steering wheel, since those are contraptions used by human drivers. The Level 5 self-driving car is not being driven by a human and nor is there an expectation that a human driver will be present in the self-driving car. It’s all on the shoulders of the AI to drive the car.
For self-driving cars less than a Level 5, there must be a human driver present in the car. The human driver is currently considered the responsible party for the acts of the car. The AI and the human driver are co-sharing the driving task. In spite of this co-sharing, the human is supposed to remain fully immersed into the driving task and be ready at all times to perform the driving task. I’ve repeatedly warned about the dangers of this co-sharing arrangement and predicted it will produce many untoward results.
For my overall framework about AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/framework-ai-self-driving-driverless-cars-big-picture/
For the levels of self-driving cars, see my article: https://aitrends.com/selfdrivingcars/richter-scale-levels-self-driving-cars/
For why AI Level 5 self-driving cars are like a moonshot, see my article: https://aitrends.com/selfdrivingcars/self-driving-car-mother-ai-projects-moonshot/
For the dangers of co-sharing the driving task, see my article: https://aitrends.com/selfdrivingcars/human-back-up-drivers-for-ai-self-driving-cars/
Let’s focus herein on the true Level 5 self-driving car. Much of the comments apply to the less than Level 5 self-driving cars too, but the fully autonomous AI self-driving car will receive the most attention in this discussion.
Here’s the usual steps involved in the AI driving task:
- Sensor data collection and interpretation
- Sensor fusion
- Virtual world model updating
- AI action planning
- Car controls command issuance
Another key aspect of AI self-driving cars is that they will be driving on our roadways in the midst of human driven cars too. There are some pundits of AI self-driving cars that continually refer to a utopian world in which there are only AI self-driving cars on the public roads. Currently there are about 250+ million conventional cars in the United States alone, and those cars are not going to magically disappear or become true Level 5 AI self-driving cars overnight.
Indeed, the use of human driven cars will last for many years, likely many decades, and the advent of AI self-driving cars will occur while there are still human driven cars on the roads. This is a crucial point since this means that the AI of self-driving cars needs to be able to contend with not just other AI self-driving cars, but also contend with human driven cars. It is easy to envision a simplistic and rather unrealistic world in which all AI self-driving cars are politely interacting with each other and being civil about roadway interactions. That’s not what is going to be happening for the foreseeable future. AI self-driving cars and human driven cars will need to be able to cope with each other.
For my article about the grand convergence that has led us to this moment in time, see: https://aitrends.com/selfdrivingcars/grand-convergence-explains-rise-self-driving-cars/
See my article about the ethical dilemmas facing AI self-driving cars: https://aitrends.com/selfdrivingcars/ethically-ambiguous-self-driving-cars/
For potential regulations about AI self-driving cars, see my article: https://aitrends.com/selfdrivingcars/assessing-federal-regulations-self-driving-cars-house-bill-passed/
For my predictions about AI self-driving cars for the 2020s, 2030s, and 2040s, see my article: https://aitrends.com/selfdrivingcars/gen-z-and-the-fate-of-ai-self-driving-cars/
Micro-Movements and the Development of AI Self-Driving Cars
Returning to the topic of micro-movements of human driving behavior, let’s consider how these actions come to play in the design and development of AI self-driving cars.
I usually get asked two questions about this topic, which comes up when I am speaking at industry conferences.
First, does it matter whether AI self-driving cars are developed such that they will be able to detect the micro-movements of other cars?
My short answer is that yes, it does matter and AI developers at the auto makers and tech firms should be including this capability into their AI self-driving cars.
Most of the auto makers and tech firms are not yet encompassing the detection of micro-movements. This lack of attention to the topic is due to several reasons.
Some of the auto makers and tech firms are oblivious to the nature of micro-movements and so it isn’t even a matter that has reached their awareness to do something about. Of those that are aware of it, they are already so taxed with trying to do straightforward driving tasks that trying to add the assessment and use of micro-movements in determining how to drive the AI self-driving car is an overload for them right now.
Indeed, many of those AI developers would contend that analyzing micro-movements and using those analyses for guiding the driving behavior of the AI is an edge or corner case. In the parlance of the AI field, anything judged to be a corner case or edge case is something that can be put on the backburner. It is not at the core of what they believe needs to be done. As such, they might consider micro-movements interesting, but not vital enough to warrant being on the pressing To-Do list.
I’d like to challenge that assumption about the low-priority assigned to dealing with micro-movements. I would argue that being able to detect and decipher the micro-movements of other drivers is a core driving skill. It allows you to anticipate what other cars and their drivers are potentially going to do. By knowing about micro-movements, seasoned human drivers can keep themselves out of potential troubles, which otherwise those human drivers that seem out-of-tune about micro-movements are more prone to getting caught off-guard by other drivers and their driving behaviors.
Some AI developers would say that detecting micro-movements won’t matter in a world of entirely AI self-driving cars, since the AI self-driving cars will presumably all be electronically communicating with each other via V2V (vehicle-to-vehicle) communications. As such, in my example of being at a red light and making a turn, if I was in an AI self-driving car it would have in-theory sent a message to the eager beaver AI self-driving car behind me, and warned it that my AI self-driving car might opt to surge ahead and yet might also then hit the brakes, all because of trying to gauge the traffic situation.
Yes, it is hopefully going to be the case that in an AI self-driving car world we’ll see that kind of V2V taking place, but meanwhile we are all living here in the real-world. As I earlier pointed out, for quite some time we are going to have a mix of human driven cars and AI self-driving cars. Likely a long time. We cannot decide now to avoid putting in place capabilities that involve dealing with human drivers. Pretending that everything will be fine once we get to an all AI self-driving car world, if we ever do, it’s living in a dream world. AI self-driving cars must be realistic and deal with a world of human driven cars.
If you take a look at many of the AI self-driving cars being tested on our roadways today, you can often figure out that they aren’t using micro-movements to their advantage. The stilted nature of the AI driving is a giveaway that the AI is not versed in detecting, analyzing, and making use of the “tell” of human drivers making micro-movements. In a sense, it is akin to a novice teenage driver that has not yet figured out how to “read” other cars and the driving behaviors of other drivers.
Now, there are some AI developers that claim they are worried that if they do include the detection and leveraging of human drivers’ micro-movements that the AI will then potentially get confused or make worse decisions.
Well, this really is more about how good or how poorly the AI development of this feature is undertaken. Sure, if you do a half-baked job of trying to leverage micro-movement detection, you might indeed have your AI self-driving car do worse than if you did things the right way. In my view, that would seem true about any of the capabilities you might opt to include into the AI system. Capabilities that are inadequately put together can be a burden and a problem, more so than an aid and a solution.
I didn’t say that adding the capability was going to be easy. In fact, I’ll readily state that it is hard.
Trying to use the sensors of an AI self-driving car to detect that the wheels of a car ahead are slightly askew, it’s not easy, and you need really good sensors and really good sensor processing software. The same can be said about detecting when nearly any of the typical micro-movements are undertaken. By definition, those movements are subtle. Picking up on the cues or tells of other cars is hard for other human drivers to do, and admittedly hard for an AI system to do.
For my article about AI developers’ overload: https://www.aitrends.com/selfdrivingcars/developer-burnout-and-ai-self-driving-cars/
For the role of internal AI naysayers, see my article: https://www.aitrends.com/selfdrivingcars/internal-naysayers-and-ai-self-driving-cars/
For how groupthink among AI developers can go awry, see: https://www.aitrends.com/selfdrivingcars/groupthink-dilemmas-for-developing-ai-self-driving-cars/
For the reverse engineering of AI self-driving cars, see my article: https://www.aitrends.com/selfdrivingcars/reverse-engineering-and-ai-self-driving-cars/
For more about edge or corner cases, see: https://www.aitrends.com/ai-insider/edge-problems-core-true-self-driving-cars-achieving-last-mile/
Training an AI Neural Network for Micro-Movement Driving Patterns
Fortunately, you can leverage Machine Learning and Deep Learning to your advantage. By training a deep or large-scale multi-layer artificial neural network with tons of collected driving related data, based on human driven cars as they are driving in regular human driving traffic, there are patterns of the micro-movement driving that can be trained for.
These trained beforehand Machine Learning and Deep Learning algorithms can then be included into the on-board AI system of the self-driving car. In many respects, this is really just a deeper form of those capabilities. The odds are that Machine Learning and Deep Learning are already being used to examine sensory data for aspects such as street signs, the presence of pedestrians, etc. The micro-movements are essentially a step deeper into that kind of analysis.
As mentioned earlier, there are some tough aspects about trying to achieve the inclusion of the micro-movement analysis and usage capability. Will the cameras provide sufficient visual detection to spot the micro-movement? Will the radar provide sufficient detection? Will the LIDAR provide sufficient detection. Also, even if you can get the sensory data, what about the added computer processing needed to go deeply into the analysis, will that be on-board? Does that add cost, weight, heat, or other concerns in terms of providing the AI self-driving car with this capability?
You also need to realize that the micro-movement is not merely about detection. Once you’ve detected a micro-movement, it needs to be populated into the AI virtual world model. The AI action planner has to be then be versed in what to do with the indications about the micro-movements. If the AI action planner ignores the micro-movement detection, the whole effort is for naught.
The AI action planner also has to deal with micro-movements that might not be a true tell per se.
Recall that I earlier mentioned that in poker playing you might be mistaken about someone’s tell. Perhaps you thought that every time they rubbed their cards with their left hand that it meant they were bluffing, but maybe that’s not a consistent tell at all. Just because a car ahead of you might be leaning toward the bike lane, it doesn’t necessarily mean that the car driver is going to bust into the bike lane.
In this case, the AI needs to be able to include the micro-movements as part of a larger picture or “understanding” of what is happening in the traffic around the AI self-driving car. You cannot in isolation try to make use of the micro-movements. When you are behind another car, you might have several minutes to observe any of their micro-movements. By giving ample time to do so, their pattern of driving behavior is likely going to be more apparent than trying to make a snap judgement.
It is also important to realize that human drivers are going to try and spoof or prank AI self-driving cars. This has already happened in instances of four-way stops, involving human drivers that “intimidated” an AI self-driving car by rolling through the stop sign. The AI was trained to wait until the other drivers came to a full stop. It was the proper way to do things. Human drivers though aren’t necessarily “proper” about how they drive. In the four-way stop, the humans figured out that the AI self-driving car would remain still until the human driven cars had waited their turn, and so the human drivers just kept moving, essentially causing the AI self-driving car to freeze in place and continuously wait for its turn to go.
I mention this example of the four-way stop to also point out that the interpretation of micro-movements of human drivers could be a means for human drivers to trick the AI self-driving car. If you as a human driver know that the AI is going to watch the alignment of your wheels, you could do a kind of basketball-like head fake and shift the direction of your wheels. It might not be due to actually needing to turn the wheels, but instead a ploy to get the AI to perhaps grant you that sought for parking spot or otherwise give you an advantage over the AI system that’s driving the self-driving car.
Overall, I am an advocate of including the detection, analysis, and leveraging of human driving micro-movements for the AI to be able to do a better job of driving a self-driving car. It isn’t going to be easy to include this capability. It will though make for a more life-like AI driving system and one that since it is going to be immersed among human drivers could be considered an essential driving skill.
When I aided my children in learning to drive, I made sure to explicitly point out the micro-movement’s aspects. I’m sure that many parents don’t think about it and aren’t aware to bring it up for their novice driver children and assume that the kids will somehow just pick it up on their own. There are some AI developers that seem to think the same might happen for AI self-driving cars, namely that rather than trying to overtly train the AI to be mindful of micro-movements, those AI developers hope or assume that it will somehow emerge by the AI’s own divinity. I doubt it will.
For more about Machine Learning, see my article: https://www.aitrends.com/selfdrivingcars/machine-learning-benchmarks-and-ai-self-driving-cars/
For more about Deep Learning, see my article: https://www.aitrends.com/selfdrivingcars/ai-machine-child-deep-learning-the-case-of-ai-self-driving-cars/
For my article about the difficulties and goal of achieving one-shot learning, see: https://www.aitrends.com/selfdrivingcars/seeking-one-shot-machine-learning-the-case-of-ai-self-driving-cars/
For the aspects of egocentric AI developers, see: https://www.aitrends.com/selfdrivingcars/egocentric-design-and-ai-self-driving-cars/
For the use of probabilities in AI self-driving cars, see my article: https://www.aitrends.com/ai-insider/probabilistic-reasoning-ai-self-driving-cars/
I’ve tried to answer the first question which was whether or not AI self-driving cars should be outfitted with a capability to detect, analyze, and leverage the micro-movements made by human drivers. Of course, I said yes.
The second question that I get asked is whether the AI self-driving car should make use of micro-movements, doing so in a similar manner to how human drivers use micro-movements.
My answer is yes, the AI ought to also make use of micro-movements. Here’s why.
Human drivers are generally used to detecting the micro-movements of other cars. I realize that I’ve also given examples of human drivers that seem oblivious to detecting micro-movements, but those are a smaller percentage of drivers.
I would also claim that many drivers might not be consciously aware that they are being affected by the micro-movements of other drivers, and yet they are indeed being impacted. You might catch out of the corner of your eye, the subtle micro-movement of another car. Doing so, you might react without any recognizable or spoken aloud reason and respond to that micro-movement, nonetheless. Humans are quite good at doing pattern matching. They often are able to find patterns when they aren’t even thinking explicitly about doing so.
An AI self-driving car should be sending the same kind of subtle “signals” to human drivers, doing akin to what other human drivers do. This again is part of my overarching belief that if AI self-driving cars are going to be driving among human drivers, which we know to be the case, those AI drivers need to be doing actions that human drivers do.
If the AI lacks the micro-movement showcase capability, it means that those human drivers around the AI self-driving car will no longer have an essential “tell” that can forecast what the AI self-driving car is going to do. Without the tell, the human drivers are going to be caught unawares. What, that stupid darned AI self-driving car made a right turn, I wasn’t expecting it, will be a common complaint.
And in a manner of speaking, I’d say that those human drivers will be right. They will be well-justified in their complaining. Who wants to have other drivers (human or AI), for which those drivers do things seemingly without the usual warnings (or tells) that we already have grown accustomed to seeing? It’s a recipe for disaster.
Would you be willing to put onto our roadways a slew of drivers that aren’t driving as is customary? Not a good move.
If you did so, it would be like dividing the traffic into two classes of drivers. When you combine with this that the human drivers won’t readily be able to gauge what the AI driver is going to do, and if the AI driver isn’t also versed in reading micro-movements and therefore unaware of what the human driver is going to do, you are setting things up for a very messy and untoward outcome.
I sometimes liken the micro-movements topic to the nature of dancing. If two people are going to dance together, they need to figure out the subtle movements that indicate whether to turn to the left or turn to the right, and whether to speed-up or slow down the pace.
Each dancer needs to be able to detect the micro-movements of the other, along with knowing how to react.
Furthermore, each dancer needs to be able to exhibit their own micro-movements, in order to allow their dancing partner to detect the movement and be able to react accordingly.
Right now, we have humans that detect the micro-movements of other human drivers, plus human drivers emit micro-movements as part of their driving efforts (by-and-large). We are in the midst of witnessing AI self-driving cars being fielded and tested that don’t detect the micro-movements of other cars, and nor do the AI self-driving cars emit their own micro-movements.
You can’t have much of a fluid dance when one of the partners is clueless. In the case of cars, a dance partner that is clueless can lead to untoward car accidents. Human drivers won’t judge what the AI is going to do, and bang, their will be collisions. AI self-driving cars won’t be able to well-judge what human drivers are going to do, and as a result the AI will drive in an overly timid and hesitant way, which will likely ultimately lead to car accidents.
It’s time to make sure that the dancing partners of human drivers and AI drivers are versed in the same kinds of capabilities and tactics of driving. I vote that the auto makers and tech firms put more effort toward the micro-movement elements, which makes sense to ensure that the dancing partners won’t collide into each other and make ghastly missteps.
Copyright 2019 Dr. Lance Eliot
This content is originally posted on AI Trends.