Anyone else concerned about fog without radar?

[JBee]

Yes, I think that's a lot of it. And I think a lot of it stems from their ability to convince themselves that they know more about a subject than someone who really does. Believe me when I say that I've dealt with this a lot in several fora dealing with various subjects. Someone makes an absurd claim. All they have to do is look up the subject on Wikipedia to see how absurd it is but they haven't done that and if you suggest they do they ignore that advice and continue to insist that they are right even if it violates a physical law or mathematical principle. It becomes pretty clear that there is no point in arguing with such a person and refraining from doing so is my general approach. The only exception is when these people say that it is OK to do something that, if done, might get someone hurt or result in a building code violation or something like that. Thus I leave it to the readership to decide whether to ignore or follow these people. To an amazing degree the readership follows them. A poll was done here I think asking who the members considered to be the most informative. There are a handful here who are astonishingly consistent in being wrong in what they post. They were rated highly as being the most informative. Thus there is no incentive for them to check on the accuracy of their postings as they are equally rewarded for posting disinformation as those who post accurately.

--

I think in trying to solve a problem that requires estimating the state of system one should consider any sensor that is likely to improve the desired estimate. There are obviously trades. Phased arrays obviously improve angular DOP but require coherent processing. That's tricky with light but it can be done thus implying expense which must be traded against the gain in the state picture one might gain from adding such a sensor. Getting back to Dunning - Kruger, I am smart enough to see the potential benefit of having such sensors in the suite but too stupid to be able to assess what that benefit might be in terms of the overall requirements.

Completely agree.

The joke is that many a mainstream media seems to follow the same strategy. It's all about hits, not content or accuracy. It then starts spreading everywhere by people who by the sheer action of clicking it validate it's spread. And then once it reaches critical mass, it becomes the de facto truth. It's quite nuts when you think about the unbridleness of it all.

The primary reason for this is that a) people have lost their BS filter in the realtime media consumption frenzy FB age and b) this is a completely unnatural platform for information distribution system, in it's capacity to amplify across the world the smallest most insignificant random opinion. Naturally, our voices can only capture an audience in earshot. Maybe a few thousand, and for that to occur you have to organise their participation first.

This more "natural" approach gave those ideas time to percolate and be tested before it left for the wider masses, passing through several gates of BS filters alone the way, with some added whispers and refinements too. It also took time to spread, so that our minds could catch up with its ramifications, before it spread to far and wide. The "good thing" was that only truly radical ideas, that were somewhat tested by one community, actually ended up leaving it for somewhere else. Now we have so much opinion, you can seemingly choose "your facts" by a simple google search, simply because like minded people can find eachother and rally together for mob behavior and validate themselves by voting with traffic clicks.

A bit like this forum. ?


---


I often have sporadic bursts of ideas in my search for amalgamation of systems.
Some are not valid, mostly because I lack the knowledge to understand their scope, but it's sometimes fun to probe the unknown and learn from curiosity. SDR used to be a bit of a pet subject of mine as we were going to use it along with CV recognition for SAR to find mobiles of lost people using our quadplane UAVs.

Sponsored

 

[tidmutt]

Completely agree that all technology is a iterative process.

But in saying that some systems are worth advancing and some are not.

From a first principles, technology as a tool standpoint, any vehicle is a movable surface that can translocate to any other point over another surface. If it does so by wheel, rail, air cushion, buoyancy etc is really dependent on it's ability to interact between the boundary layers to create propulsive motion in the desired direction.

FSD is the result of too many degrees of freedom, most of which are unnecessary, and produce their own huge subset of problems and resource requirements.

The reason is that the basic "idea" of a car is loaded with hereditary systematic disease. Where the systems are derivatives of other inventions that do not assist the primary function, but rather address the required external interactions (like indicators for traffic interaction) with even more undesirable effects (window visibility at intersections etc).

Most of these can, and have be solved with a rail system. How many trains have you seen with FSD over the last 300 years? All of them have set and forget throttles and don't come with a steering wheel even as an option!

And I'm not kidding that I think cars are dumb. I can still like driving them and buying them though. I do dumb things all the time for fun. Yes there are use cases where rail won't work, but for 99% of human traffic a combination of "pod" rail and evtols is better, much safer, faster, less resource intense, easier to automate, is therefore more affordable and has less environmental impact.

This is a good example of how autonomous personal rail could be done if it had electric assist:

Right, but we aren't going to eliminate cars. I think Tesla did what they had to do.

Every city should likely be constructed around mass transit, we shouldn't live in giant houses that sprawl out from cities. We should live in higher density dwellings, walk, bike, scooter to work and play. Using mass transit for long journeys. Perhaps some other form of personal transportation. In many cases we don't though... kind of have to work with what we have.

 

[Crissa]

At least we should build enough high-density stuff! So that it's not more expensive than sprawl for those who want to live there.

Sprawl costs more taxes for roads, wires, sewers, time. It's just, the cost is less visible as it consumes little minutes of your day, car costs, etc costs. It's an invisible tax on everyone, especially the young, old, poor, or disabled.

-Crissa

 

[jhogan2424]

At least we should build enough high-density stuff! So that it's not more expensive than sprawl for those who want to live there.

Sprawl costs more taxes for roads, wires, sewers, time. It's just, the cost is less visible as it consumes little minutes of your day, car costs, etc costs. It's an invisible tax on everyone, especially the young, old, poor, or disabled.

-Crissa

That’s one reason I love where I am in rural Arkansas. 500k will buy you a nice 3,000 foot home on 40 acres all to yourself (with triple garage!), a pond, a boat, a barn, an ATV, a workshop, a tractor to maintain everything with, and a Cybertruck. Most importantly, all of this located with 100,000 pine trees between you and your closest neighbors. You will have to pay for your own road, utility access, and septic system though, we don’t give those away here.

 

[ajdelange]

SDR used to be a bit of a pet subject of mine ...

Ah, the nostalgia. That was nearly 40 yrs ago when we started fiddling with "software radios" as we called them then. FORTRAN on a VAX with an attached array processor and then someone tried to build one with 2 MHz processing bandwidth with about 8 array processors (a good sized roomful). They clocked at 11 MHz! Now all that stuff (and more) is in your iPhone.

 

[FutureBoy]

OK, OK... I'm late to the party on this thread. I kept thinking it would die out and I could just ignore it. But alas, that has not been the case. So I just need to get this out of my system.

Regarding the thread title:
Anyone else concerned about fog without radar?

My response:
No. I'm more worried about radar without fog.

I get that the radar seems like a good idea. Even Tesla thought it was a good idea and was using it in their cars for a long time. But from various AI-themed talks I've seen regarding this topic, it seems that the AI was getting limited by the relatively imprecise radar data input. This was causing the visual inputs to no longer discern at the level of detail needed for FSD. So the radar functionality was turned off and suddenly the visual inputs were starting to learn again.

I'm not an AI expert. I wasn't in the Tesla meetings or AI team. But from the public talks that AI team members have given, that is what I understood.

Given all that, I'm more concerned with getting the full FSD system up to the necessary abilities for general availability than I am for some edge cases. Soupy fog is an edge case that even human drivers probably shouldn't even be driving in. I get that it happens more often in certain locations but if you can't actually see, it isn't safe to drive.

Now it wouldn't surprise me if sometime in the future after FSD is working well and probably widely available, Tesla might find a way to integrate some other sensing mechanism that would allow for more accurate driving in fog or other hazardous conditions. At the moment though, I'm more concerned with getting FSD out and generally available so that it can save lives and provide better living experiences for people. The inconvenience of foggy weather does not outweigh the daily death toll of regular humans driving cars.

Anyway, not that I got that out of my system...

Back to your regularly scheduled programming.

 

[jhogan2424]

OK, OK... I'm late to the party on this thread. I kept thinking it would die out and I could just ignore it. But alas, that has not been the case. So I just need to get this out of my system.

Regarding the thread title:
Anyone else concerned about fog without radar?

My response:
No. I'm more worried about radar without fog.

Back to your regularly scheduled progr

Ok, so you’re not worried about fog. Are you worried about the car stopped in the road in front of the car in front of you that is playing on their phone?

 

[jhogan2424]

Why did I even reply to that? I promised myself I was finished with this one. ?

 

[FutureBoy]

Why did I even reply to that? I promised myself I was finished with this one. ?

Just as a note, you are able to edit or delete messages you wrote. Although, if you do it too late, someone could quote you first.

 

[FutureBoy]

Ok, so you’re not worried about fog. Are you worried about the car stopped in the road in front of the car in front of you that is playing on their phone?

Actually I’m not because I already have that same problem when I drive using my own eyes. As long as FSD can process fast enough to recognize and respond faster than my nervous system it will end up being a better and safer driver than I am.

 

[jhogan2424]

As long as FSD can process fast enough to recognize and respond

Well, it can’t, if the photons from the stopped vehicle have never reached the camera lenses. Please look back in this thread at the video posted where radar saved the day for one family, if not more. There are some comments suggesting the cameras may have picked this up and caused the car to stop but it was not the cameras, it was the radar. Forget the fog scenario and just think about the stopped vehicles that CAN NOT BE SEEN by any camera supplying data to any AI system. Do yourself a favor and disregard all comments including mine and just watch the video for yourself using common sense. That is all I have to say on this subject forever more, no matter how luring the responses.

 

[FutureBoy]

Well, it can’t, if the photons from the stopped vehicle are not reaching the camera lenses. Please look back in this thread at the video posted where radar saved the day for one family, if not more. There are some comments suggesting the cameras may have picked this up and caused the car to stop but it was not the cameras, it was the radar. Do yourself a favor and disregard all comments including mine and just watch the video for yourself using common sense. That is all I have to say on this subject forever more, no matter how luring the responses.

The one common sense thing about the scenario is that if the photons can’t reach the camera then they cannot reach my eyes either. So as long as the FSD processing is faster at reacting than I am, it’s going to be safer than having me drive.

And FSD doesn’t get distracted or tired so in those scenarios it definitely has me beat.

 

[jhogan2424]

??????????????
Everyone is entitled to opinions but give me a break here. This is a VIDEO that shows what happened. It’s not an opinion, it’s a fact. I’m not trying anymore to explain something that’s so damn simple. I think some of you just want to argue or else you have not watched the video!

 

[jhogan2424]

I have only read a few posts here, sorry if I am being redundant. FSD hopes to get to the level of a competent human driver. How are we able to drive in fog without radar now? We use vision and go slowly or pull over if it is unsafe. Ai should consistently make the right call to stop when it is unsafe, human drivers do not always do that.

You’d do well to read them all, and watch the video. I’m no longer trying to explain it, only referring argumentative types to the video that speaks for itself 100%.

 

[John K]

??????????????
Everyone is entitled to opinions but give me a break here. This is a VIDEO that shows what happened. It’s not an opinion, it’s a fact. I’m not trying anymore to explain something that’s so damn simple. I think some of you just want to argue or else you have not watched the video!

There are two assumptions being made

1. The video shows Tesla’s radar identified the accident was about to happen. The assumption is challenge due reviewing the video, I could see issue in the video before the warning showed before the brake. The video does not illustrate if the radar or visual cameras were first to detect.
2. Assumption radar is the only method to see cars ahead of the cars directly ahead. I could see cars ahead in the video and do so in my normal driving. If I can, the visual camera can too by utilizing multiple cameras view and perspective to identify items around.

if you cannot live without radar, cancel your order or petition Tesla. I can live without radar in Tesla vehicles due to some knowledge on the matter and company statements. I trust the system will work safely.

I am not stating never radar, nor never lidar, those technology capabilities, IMO, do not perform at the required level and some are cost prohibitive.

Another option, dont invest in FSD and rely on your driving ability.

Sponsored