Those Fish That Rained Down From The Skies On A Texas Town Was Not A Fishy Tale (It Happened), Netting Some Quite Catchy Insights For AI Self-Driving Cars

Sometimes, when it rains, it pours.

Those raindrops keep falling on your head, though if you are inside your car the odds are that the rain is instead pounding fiercely down upon the automobile rooftop and dangerously drenching the roadway ahead of you. Driving in the rain can be plenty scary. Some people pretend that it is no big deal, though the reality is that the driving world is a much more hazardous place while amidst any modicum of a raining torrent.

Astute drivers in rainstorms try to prepare themselves for the unexpected.

For example, other cars can suddenly swerve without any due notice. Those drivers might not have realized that the roadway was slick due to an insidious and somewhat undetectable layer of rainwater. One moment they are driving as though they didn’t have a care in the world, and the next moment their vehicle began playing a serious and potentially life-threatening gambit of injurious slip and slide. I’ve covered the challenges of rain-related driving previously in my column, such as the analysis at this link here.

Another example of rain-related gotchas consists of an animal deciding that it wants to dart across the street, directly in front of your in-motion car. Perhaps a neighborhood dog is loose and wants to get out of the rain, figuring that some warm and protected hutch on the other side of the road provides suitable cover. Or in a rural area, there is always the chance of a deer sprinting swiftly across the path of your vehicle.

When an animal provides that kind of a surprise, you would normally on a dry and sunny day be able to jam on your brakes, or possibly make a momentous swerving action to avoid the living creature. The rain tosses those options somewhat into the air. Your ability to bring the car to a rapid halt is undoubtedly nixed by the slickness of the roadway. Making a swerving action is going to be likely out of control and you could end up in a ditch or maybe even roll the car. For those of you especially interested in such aspects about coping with animals that without advanced warning opt to cross the road (including chickens), see my coverage at the link here.

We all know that driving in the rain requires being extra cautious.

The usual rule of thumb is to drive more slowly than usual. Keep your eyes wide open. Ensure that your windshield wipers are in good shape and able to shove the dousing rainwater out of the way of your eyesight. Have tires that contain ample tread and have sufficient air in them. Anticipate that other drivers will be either idiots or decidedly reckless in how they opt to drive while in a rainstorm. Remember that your means of controlling the car is reduced, requiring extra distances and being mindful of overplaying the brakes or the steering.

As I say, drivers that are on the ball will anticipate the unexpected.

Allow me to provide an instance of the unexpected that is truly the unexpected.

Are you ready?

Envision that fish were falling from the sky, doing so amongst the rain pouring down.

Yes, I said fish.

Last week, a town in Texas had the veritable once-in-a-lifetime chance of witnessing and driving in a spectacle that the news referred to as so-called “fish rain” (that’s assuredly different from everyday hard rain or even good old-fashioned annoying rain).

The belief is that the fish were scooped up by giant waterspouts that occurred in the surrounding area. For a brief moment in time, these flightless fish were certifiably flying fish. The waterspouts readily lifted the fish up into the sky and then redeposited the slimy, shiny, delicately flopping aquatic gill-bearing creatures here and there.

I realize that many of you have undoubtedly loudly exhorted at some time in your lives that you’ve been in exceedingly foul weather where it was ostensibly raining cats and dogs, though, of course, no cats or dogs were actually involved. The people of Texarkana, Texas can now do some one-upmanship and attest to the fact that they experienced raining fish.

You would seemingly have to agree with me that this is a truly unexpected instance of unexpectedness.

That being said, there is indeed a phenomenon known as animal rain. Per National Geographic, animal rain is a documented meteorological fact. Throughout history, there have been documented and at times undocumented claims of flightless animals falling from the sky. Rain might or might not be involved. Generally, the conventional means by which this happens entails waterspouts that perchance grab up whatever might be around, the objects are transported over some random distance, and eventually, the weight of the objects overcomes the lifting forces.

Ergo, animals fall from the sky.

You might be surprised to know that the animals can sometimes be carried miles away from their original pick-up point. A wild ride ensues, that’s for sure. The animals are mostly smaller creatures such as fish or frogs.

As an aside, here’s a puzzler for you and something you can ask a friend about too if you want to engage in a hopefully civilized debate on a topic of admittedly somewhat valueless merit.

Which is more preferable, namely having fish fall from the sky or having frogs fall from the sky?

I suppose anyone that likes to go fishing might vote for the fish falling from the sky. This seems like a much more expedient method of netting some fish. But there is another side to that coin. If the fish are being taken from your favored lake or pond that you relish fishing at, the removal of the fish by the waterspout is going to undercut your next fishing trek.

The whole matter is a dizzying conundrum.

Speaking of being dizzy, remember that we were talking about driving in the rain and the overall theme was to be expecting the unexpected. Is it reasonable to anticipate that fish or perhaps frogs would fall from the sky during a rainstorm?

I don’t think we can judge ourselves as being narrow-minded or faulty in our thinking for not anticipating that kind of rain.

There you are in your car. The rain is coming down strongly. Your windshield wiper blades are trying mightily to brush aside the sheets of water. You can barely see out the windshield. Because you are a savvy driver, you are going more slowly than normal and sitting on the edge of the driver’s seat, awaiting any kind of indication that something in the traffic or on the roadway is amiss.

Plunk, a fish plops onto the hood of your car.

Bang, another fish bounces off the rooftop of your rain-soaked vehicle.

As you stare ahead at the slick street in front of you, there are fish landing one after another onto the puddles on the roadway.

Seems to me that you would have to blink your eyes a couple of times and remind yourself that you are awake, you are driving your car, it is raining, and you are immersed in reality. I emphasize this notion since these days you might somehow be in the metaverse or have miraculously been transported into the matrix (and not know you are beyond the bounds of conventional reality).

You never know.

Anyway, back to real reality. After getting over the shock of the raining fish (can you really get over it?), hopefully, you would take the matter in stride. By this, I mean that you would continue to safely drive. No need to panic. Any panicking could make the situation even worse.

A sizable temptation would be to try and steer around the fish that are now residing on the roadway. Nobody wants to drive over an innocent school of fish that so happened to get the journey of their lifetime. On the other hand, radical driving actions to avoid the fish could be problematic. You might go off the roadway or ram into other nearby cars.

One worry would be that a fish manages to smack your windshield. Would the windshield shatter or remain intact? Your thoughts wander to the assumption that the automotive manufacturer dutifully tested their windshields and devised the glass to withstand the striking force of a fish. They must have figured this out, you stridently tell yourself.

Okay, the bottom line is that you probably will survive the fish-laden rainstorm. Assuming that other drivers do not go berserk, you will drive along and squish some here or there, but otherwise make your way to wherever you are going, unscathed. 

Will the insurance company believe you when you submit a claim to cover the damage from the fish that dented your hood and rooftop?

Those darned insurance companies often seem to find the lamest of excuses to deny claims. The insurance adjuster would laughingly tell you that your story is a shameful fish tale and ask you to fess up to the real truth of the matter. I guess that particular adjuster has seen a thing or two, but not flightless flying fish.

Here’s a notable idea. Better make sure to keep the automotive smashing fish handy in case you need to showcase evidence of what struck your beloved car.

Now that we’ve covered the raining fish, we can consider a somewhat intriguing and quite different perspective on the topic.

Here’s a noteworthy question that is worth pondering: What would an AI-based true self-driving car do in a rainstorm that included those pesky raining fish?

I’m glad you asked.

Allow me a moment to unpack the question.

First, note that there isn’t a human driver involved in a true self-driving car. Keep in mind that true self-driving cars are driven via an AI driving system. There isn’t a need for a human driver at the wheel, and nor is there a provision for a human to drive the vehicle. For my extensive and ongoing coverage of Autonomous Vehicles (AVs) and especially self-driving cars, see the link here.

I’d like to further clarify what is meant when I refer to true self-driving cars.

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 Level 4 and Level 5 (see my explanation at this link here), 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, see my coverage at this link here).

Since semi-autonomous cars require a human driver, the adoption of those types of cars won’t be markedly different than 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.

Self-Driving Cars And The Potential For Animal Rain

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 is 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.

With that clarification, you can envision that the AI driving system won’t natively somehow “know” about the facets of driving. Driving and all that it entails will need to be programmed as part of the hardware and software of the self-driving car.

Let’s dive into the myriad of aspects that come to play on this topic.

First, it is important to realize that not all AI self-driving cars are the same. Each automaker and self-driving tech firm is taking its own approach to devising self-driving cars. As such, it is difficult to make sweeping statements about what AI driving systems will do or not do.

Furthermore, whenever stating that an AI driving system doesn’t do some particular thing, this can, later on, be overtaken by developers that in fact program the computer to do that very thing. Step by step, AI driving systems are being gradually improved and extended. An existing limitation today might no longer exist in a future iteration or version of the system.

I trust that provides a sufficient litany of caveats to underlie what I am about to relate.

We are primed now to do a deep dive into how self-driving cars might cope with raining fish.

I’ll start with the rather extreme viewpoint proffered by smarmy skeptics that no self-respecting self-driving car would ever be caught in the midst of a fish raining rainstorm. I bring this up because there are some pundits that keep insisting that self-driving cars will never ever get into car crashes of any kind.

One supposes that this includes the crashing actions of fish falling from the sky.

You see, the pundits espousing this “never going to crash” mantra are seemingly mesmerized with the fanciful idea that the AI of self-driving cars is going to be so darned good that no adverse situation will catch the AI off-guard. The AI is amazingly all-knowing and all-seeing. Sorry, that’s hogwash. Worse still, it is setting utterly preposterous expectations. For more of my analysis about this disconcerting bad-to-the-bone knucklehead idea of self-driving cars never crashing, see the link here.

Moving on, I trust that you’ll go along with me that there is a reasonable chance that a self-driving car could end up in a setting of animal rain.

Let me clarify that the odds of animal rain are extremely slim, to begin with. As such, it is fair to suggest that very seldom if ever will a self-driving car encounter animal rain. In the same light, it is fair to equally suggest that very few human drivers will ever encounter animal rain. It just doesn’t happen with sufficient frequency to statistically be plausible for everyday driving purposes.

But rarity does not equal to never.

It can in fact happen.

Should the automakers and AI self-driving car developers be worried about programming their self-driving cars to cope with animal rain?

I would dare say that amongst the many commonly occurring dangers and oddities that can befall driving on our roadways, animal rain such as falling fish does not deserve to be high on the priority list. If there was ever something that justifiably goes onto the so-called edge cases list, animal rain takes the cake.

An edge case is considered an instance of a driving circumstance that is not within the considered core of driving activities or encounters. Those devising self-driving cars cannot cover everything that needs to be done in an all-at-once mode. Instead, lists of features and functionality are stratified from high priority to low priority.

Presumably, the focus goes to the higher priority items first.

Thus, you can assume that right now there are pretty much no AI-based true self-driving cars that have been programmed to specifically account for driving in a storm of flightless flying fish. Nor amid flightless flying frogs. And so on.

This is a significant point pertaining to what a self-driving car would do in the oddball animal rain occurrence.

The overall generalized programming that deals with the driving of the autonomous vehicle would need to handle the situation. There isn’t any specialized code or dedicated subroutine sitting within the AI-based driving system and awaiting its moment to shine by being invoked as a result of encountering the rain fish.

We can dreamily speculate about what a dedicated piece of AI software might do to contend with the animal rain. I’ll let the automakers and AI developers stew on that, no need to freely hand them the requirements and specifications.

If you are wondering whether the vaunted use of Machine Learning (ML) or Deep Learning (DL) might be the savior for handling the animal rain, please think carefully about that. The answer is generally a big fat no.

Here’s the deal.

Machine Learning and Deep Learning are essentially computational pattern matching techniques and technologies. The usual approach is that you collect lots of representative data that contains whatever hidden patterns you hope to have the algorithms discover and let the ML/DL churn away to try and uncover statistically valid patterns.

For example, you might collect zillions of pictures of Stop signs. You feed those pictures into a prepared ML/DL. The computer cranks through the data and seems to have a pattern matched to a high level of certainty. One would assume that perhaps the common pattern of a Stop sign being red is captured, and the shape of the Stop sign is captured, and the word “Stop” is captured, etc. All of those are pattern-related clues to try and discern what a Stop sign looks like.

Now then, if I don’t happen to have many pictures of Stop signs, it will be challenging to train the ML/DL on what a Stop sign is. This is a significant problem right now facing the ongoing and growing use of Machine Learning and Deep Learning, i.e., the lack of representative data for whatever training is desirous of being undertaken.

Researchers are aiming to devise a variant of pattern matching that can function reasonably well in what is considered few-shot or one-shot settings (see my explanation at the link here). This could alleviate some of the data spareness concerns (maybe).

Given that quick elucidation about how ML/DL works, think closely about how much-collected data there is about animal rain (in terms of actual instances).

Not much (if any) exists.

Of course, you could create computer-based simulations to concoct such data. One supposes you could even make arrangements at one of the special closed-track courses that are used to develop and test self-driving cars. I’d welcome seeing that happen. The track administrators would presumably arrange to rocket fish and frogs into the air so that they will land on the being tested self-driving cars. A sight to behold.

Back here on earth, we can make the straightforward assumption that due to the low likelihood of animal rain, and the lack of collected data, and the general fact that the odds of encountering such an oddball situation is extraordinarily remote, nobody is sparing the time to concoct AI software for this specific use case.

What then would the otherwise generalized AI driving system do?

First, the key is whether the sensors of the self-driving car would even detect the flightless flying fish.

Most self-driving cars are being armed with video cameras, radar, LIDAR, thermal imagining, ultrasonic, and other such sensors (well, not everyone, such as Elon Musk and his Tesla cars are being outfitted with a much less robust varied sensor suite per se, unlike most of the rest of the industry, see my coverage at the link here).

The sensors are almost certainly going to observe the fish, in terms of objects being detected outside of the autonomous vehicle, but the open question is whether the sensory data is going to be properly interpreted by the AI software. If there isn’t any particular programming associated with discerning the objects, the AI driving system might not have any basis for distinguishing that the fish are there.

Thus, the sensory captured data indicates that objects are falling from the sky. Let’s envision that rain is also falling from the sky.

Would the AI software be able to interpret that the falling items are both rain and fish?

Or would the fish be simply blobs that are uninterpretable or that are altogether left uninterpreted?

Conclusion

I’m guessing you still want to definitively know, well, what in the heck would an AI-based true self-driving car do?

I certainly do not want to disappoint you on that score, so let’s add up the whole kit and caboodle. Given all the aforementioned caveats, the best bet right now is that the AI would essentially “ignore” the flightless flying fish and simply continue driving ahead.

Unless the fish were large enough to register as a danger, such as say a fifty-pound giant catfish or its brethren, they would (tongue in cheek) otherwise all be flying under the radar (that’s got to be a suitably humorous and yet mildly cringeworthy quip).

The raining fish in Texas were reportedly smaller-sized fish, which also makes sense in that we wouldn’t likely expect giant fish to be scooped up, though the strength of a massive waterspout should not be underestimated (think of flying cows, but that’s a tornado movie motif). Remember, everything in Texas is larger, so prepare yourself for the day that they have marlins being scooped up and dropped onto their expansive highways.

Another way in which the AI might get wind of something amiss would be if the fish plummeted down on some of the sensors of the autonomous vehicle and accordingly broke the sensors into smithereens. When a crucial sensor gets struck or demolished, the AI usually is programmed to go into a safety mode of slowing down and coming to a demonstrative halt.

I hope this then quenches your thirst about what would an AI-based true self-driving car do in the face of those wingless flying fish.

Let’s pause for a reflective moment and mull somewhat about the future.

I am patiently awaiting a news story about sharks that are flying in the air and landing on the top of cars traveling along our highways and byways. When that happens, I’ll make sure to cover it and indicate what self-driving cars would do.  In case you question the chances of sharks flying around like that, I assure you that there are plenty of carefully scientifically documented movies, video games, and comic books about Sharknado’s that clearly establishes the stark reality of such a possibility.

Whatever you do, I believe you are best urged to not look up. Yes, that’s the ticket, please do not look up, no matter what is flying above you.

I heard that somewhere as a life-saving piece of advice.