How Does Steering Assist Detect One's Hands?

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I must say, I do find it curious that some people are so apprehensive about automation making its way into cars... But seemingly, they have no issue with automation being present in the aviation industry.

Automation is not necessarily a bad thing, when used responsibly.
I'd be fine with automation in road vehicles if they kept the same separation used in aviation.
A plane that hits an air pocket and drops a few hundred feet or more, is no problem. a vehicle that suddenly veers a few hundred feet is a whole different story.
When feet matter (take off and landing) the aircraft is under human control.
I am fine with some automation like in rail, where there is a physical limit to how far off track (pun intended) it can go.
Self driving semi's (articulated lorries) still can't back a trailer into a dock. That is done with a human.
Yeah, I have real reservations on the push for automation. Doing fine 90% of the time isn't good enough.
I freely admit some humans can't achieve that either, but that doesn't make it an acceptable replacement.
 
Last edited:
Whilst I don't disagree, you're not looking at "the big picture"... Eventually - likely sooner rather than later - we'll get to a point where automation is so effective that in 99% of cases, we won't need to worry about "drivers" paying little or no attention to their vehicle.

Doing fine 90% of the time isn't good enough.

I'm typing on a Windows laptop - pretty basic and well-established tech. It operates perfectly for 10-12 hours per day for weeks or months at a time, but every once in a while an update causes it to go haywire and I have to waste time reverting to a previous backup and trying to figure out what went wrong.

Any automated car is going to be vastly more complex than my laptop, with multiple sensor arrays to detect and monitor other vehicles, pedestrians, wildlife, road markings, etc while speeding down (in the US) horrible bumpy roads and often in adverse weather conditions. While it might work fine for days and weeks and months, when the 1% situation occurs and the automated system has a hiccup I will face major repercussions, as will other road users around me. That's the "big picture" I'm looking at.
 
it gets confused sometimes... Maybe the roads in my area just have substandard lane markings.
Mine has been pretty reliable, except where there's road works or a poor-condition road, then it sometimes gets confused... Especially when you have those black lines in the tarmac (where they've resealed the road?) or white lines that they've tried to paint over with black paint. Intersections or sections of the road where it changes into lots of different lanes also seem to confuse it; but I find that the system automatically deactivates in these situations, if I have not already deactivated it myself.
 
I'm typing on a Windows laptop - pretty basic and well-established tech. It operates perfectly for 10-12 hours per day for weeks or months at a time, but every once in a while an update causes it to go haywire and I have to waste time reverting to a previous backup and trying to figure out what went wrong.

Any automated car is going to be vastly more complex than my laptop, with multiple sensor arrays to detect and monitor other vehicles, pedestrians, wildlife, road markings, etc while speeding down (in the US) horrible bumpy roads and often in adverse weather conditions. While it might work fine for days and weeks and months, when the 1% situation occurs and the automated system has a hiccup I will face major repercussions, as will other road users around me. That's the "big picture" I'm looking at.
You don't want to read about the quality of the softwares that run it all.
 
You don't want to read about the quality of the softwares that run it all.
You know cars aren't running Microsoft Windows-based operating systems, right?

For a start, from a stability and security standpoint, Windows-based operating systems are about as bad as it gets... Which is why it is rarely used in environments where safety or security is paramount.

Cars are using a real-time operating system (RTOS), from a very short list of RTOS, many of which have been used in safety-critical environments for decades.

It's also worth pointing out, again, that much of the technology behind vehicle automation has been used in the aviation and other industries for quite literally decades... So this is not new technology.
 
It's also worth pointing out, again, that much of the technology behind vehicle automation has been used in the aviation and other industries for quite literally decades... So this is not new technology.
It is also worth pointing out that in aviation, the separation of aircraft is measured in miles, in automotive it is feet, there are three axis to used to avoid collisions, not two, and the operating space as far less crowded.
There really is no comparing the two environments.
 
It is also worth pointing out that in aviation, the separation of aircraft is measured in miles, in automotive it is feet, there are three axis to used to avoid collisions, not two, and the operating space as far less crowded.
There really is no comparing the two environments.
Exactly. I used to fly a lot for my work. I spent a lot of time staring out the window and regularly saw other planes that seemed quite close to the one I was in, but the "quite close" was almost certainly greater than a mile horizontally and thousands of feet vertically.

If something goes wrong with the controls or an automated system, the pilots have both time and space to avoid a collision with another plane or the ground.

On the highway, most people don't give a second thought to oncoming traffic passing by with 3-4 feet of horizontal separation at a closing speed regularly >100 mph, or to passing or being passed by an 80,000 lb semi with 2-3 feet of horizontal separation.

Plus, when I drive our Leaf or our other car (Honda Fit), the bumpers of the most common vehicles on the road -- SUVs and trucks -- sit at the level of my face.
 
There really is no comparing the two environments.

Aviation is not the only environment where this tech and RTOS have existed for decades... Elevators, heavy machinery, the rail and shipping industry, the gas and oil industry, other elements of the aviation industry (such as ATC), the space industry.

This is absolutely comparable.

If something goes wrong with the controls or an automated system, the pilots have both time and space to avoid a collision with another plane or the ground.
You think elevators, heavy machinery and the rail industry have time and space to correct problems that occur? Or that the space, gas and oil industry have the luxury of time when their sensors or RTOS play up? 😂

I have no doubt in my mind that if I took the time to research these individual industries, I could come up with dozens and dozens of Real World examples where comparable sensors and RTOS make split-second decisions about safety with little or no physical space to play with... But I just don't care enough to go proving the obvious. 🙄

Oh, and let's not forget that countless countries have tested or are currently using public transport systems which are completely automated - many of which operating at high speeds in close proximity to other fast-moving vehicles, pedestrians and / or other hazards which would present minimal notice to a human.

Even in my own city we had such a system for a while and the only reason they got rid of it (I worked for the company operating it at the time) was because people would go out of their way to try and trick the sensors of the automated EV buses, because of the negativity surrounding EVs and automation - which for the record, rarely worked, but caused all sorts of traffic, legal and logistical nightmares. So they decided it was easier to cut it away for now, until attitudes towards EVs and automation changed dramatically.

But we still have other places in Australia operating high speed public transport systems autonomously in close proximity to other vehicles, pedestrians and / or other hazards; as do LOADS of other countries around the world.

You're also ignoring the fact that many (not all) EVs are "looking" several hundred meters in each direction around the vehicle... Only the other day I watched a tech demo video of a Nissan Leaf demonstrating (full) automation, and it was identifying potential hazards as much as 1KM away, and taking the necessary precautions / getting ready to react accordingly. There are also videos online of other EVs looking even further into the distance, to identify potential hazards.

Is vehicle automation perfect right now? Hell no, and it will probably never be "perfect".

Is it ready to roll out en mass to the the General Public and not cause me to lose sleep at night? Absolutely not.

But it's pretty good right now (particularly when it comes to certain manufacturers) and just like the other bazillion industries with similar tech, operating systems and sensors, it will continue to improve, even when it hits reliability rates above 95%...

You're all scared of progress, when in reality, there's no need to be.
 
I am not scared of progress, I am scared of people trying to defeat the safety measures in place because the the system is not at present, capable preforming the task at hand.
So far it has proved less than up to the standard needed. It is further complicated that rather then be used as safety assist, it is being pushed and used to replace the drivers attention to the task at hand.
High speed rail, is highly automated, but also isolated separate tracks, no at grade crossings, fenced in.
I have seen tests done where obstacles (people, cars, animals) were either not recognized or were but then were "Lost".
The tech is not ready, which is why there are requirements for the driver to be in command, those, like you who are trying to defeat the safety measures are NOT moving the automation forward, but backwards.
As long as crashes involving these systems continue, public acceptance will not be forthcoming.
Not only does the system have to perceive potential threats (collisions) but must be able to place a hierarchical priority on the threat, a baby must be a higher priority to evade over a ball or even another vehicle.
Until it can do that, a human must be attentive and ready to take over in a fraction of a sec.
It is not just recognizing threats, something it is currently failing at, but must also be able to prioritize them. We still have Tesla's running into parked firetrucks, running into them at high speed, obviously the "driver" was not ready to take over.
We may get there some day, when that day comes, the mfg will have to accept responsibility for the vehicle actions, something they currently are unwilling to do.
 
I'm not interested in getting into an argument. I will just respectfully suggest that many of your examples are not comparable to automobiles, or even aircraft.

Elevators have built-in safety locks. If something goes wrong (e.g., weird sensor readings), the elevator freezes in place. More importantly for our discussion, if something goes wrong the elevator isn't going to swerve sideways into the next building.

Similarly, trains are on tracks, and generally do a pretty good job of staying on the tracks. Except for road crossings and certain urban areas, there's usually a buffer zone around the tracks so again, if a train comes off the tracks there's a good chance it's not going to hit anything.

Spacecraft make an interesting example, in that both hardware and software performance and reliability is critical, except when it isn't. It's certainly very important during launch, and for putting the craft in a specific orbit or, for example, making a successful landing on Mars. But, once an object is in space, it's surrounded by... almost infinitely empty space. The Voyager 1 and 2 probes were launched 47 years ago. They're currently 15.5 and 12.9 billion miles from Earth, respectively. Pretty amazing, yet they regularly have both hardware and software issues that complicate communication with Earth and could end the mission (not to mention the fact that radio communication with Voyager 1 takes about 45 hours roundtrip). Luckily, since they're in space there's not much for them to run into, providing lots of time for the mission controllers on Earth to keep coming up with cool new tricks to keep them operational.

I consider driving to be an important responsibility. It's certainly the most dangerous thing I do on a regular basis. I've seen the results of too many accidents, often caused by negligence, to feel otherwise. Someday there may be automated systems that can consistently and reliably perform better than a human. I don't think that time has come yet.

It's fine if you're all for automated systems, but recognize that if you're trying to circumvent the manufacturer-designated parameters of the automated system in your vehicle (e.g., "how does my Nissan Leaf know when my hands are on the steering wheel?"), you are not being the safest driver you can be.
 
You know cars aren't running Microsoft Windows-based operating systems, right?

For a start, from a stability and security standpoint, Windows-based operating systems are about as bad as it gets... Which is why it is rarely used in environments where safety or security is paramount.

Cars are using a real-time operating system (RTOS), from a very short list of RTOS, many of which have been used in safety-critical environments for decades.

It's also worth pointing out, again, that much of the technology behind vehicle automation has been used in the aviation and other industries for quite literally decades... So this is not new technology.
Everything I’ve read over the years about vehicle software doesn’t inspire confidence.
 
Back
Top