Radar removal improved accuracy rather than harming it, they don't actually sell a car at the price point you listed. and inference from Tesla's award winning safety records suggests that Telsa does care about human life.

The factual content in your comment is decisively lacking.

Careless of you, I think.

"Radar removal improved accuracy rather than harming it..."

Reference, please.

When more sensors don't perform better than less sensors, either your hardware or your software is flawed. Removing a sensor from design because your system doesn't work speaks poorly of your product.

Even following that simplistic logic, if autopilot isn't safe, How could they dare not remove it?

> When more sensors don't perform better than less sensors, either your hardware or your software is flawed.

Consider that 99 cameras plus one radar does not make a system perform better if the task is identifying whether there is the color blue in an image. If we add ten thousand radars to this situation it doesn't produce an improvement, in fact, it gets worse, because there is more power being used by the sensors and a lot more complexity.

> Removing a sensor from design because your system doesn't work speaks poorly of your product.

This is a convoluted way of saying that making improvements to things is bad. That is nonsensical.

> Even following that simplistic logic, if autopilot isn't safe, How could they dare not remove it?

They did remove it.

Just to show you mathematically that they're right consider that if you have two sensors that are both measuring the same thing. One is Norm(mean=0, std=0.00001) and the other is Norm(mean=0, std=10). The true quantity is always zero in the real world. Given this scenario and these two senors you get the result of 0.11 and 11 as your return from the two sensors. Lets say we average the result of these two sensors. The average is farther from the true value than the first sensor was. The less reliable sensor is adding noise to our measurement. It isn't high utility.

A crafty person here will draw on ideas like kalman filters and sensor fusion to notice that as the number of measurements increase you can gain precision even in the presence of an unreliable sensor.

This is very true and it might be where the bad intuition that having multiple unreliable sensors is good is coming from. Multiple noisy readings can be averaged to get a more reliable reading if the distribution of reading errors is normally distributed. You can take ten Norm(0, 10) and average them and the result of that will be a normal distribution with a standard deviation which is lower than the standard deviation of the distribution that we sampled from. See basic statistics for the e

There are a few reasons why this doesn't apply in this case:

1. The measurements aren't going to be of the same thing. At time step t, the state of the world is potentially different from t+1. Since we have one radar sensor averaging the results of the two sensor readings doesn't give us the true position. Kalman filters solve this with a dynamics function so this problem isn't something that is impossible to solve.

2. So lets say we solve it. Even if we did measure the same thing we have another problem. We don't have time to take advantage of the central limit theorem. Yes, multiple noisy readings of the same thing average to the true reading, but since we have only a split second to decide whether to slam on the break to avoid an accident we don't have the capacity to await convergence of the noisy sensor and noisy dynamic function to the true measurement. But lets say we did even though we don't.

3. The radar sensor isn't actually a normal distribution in terms of its noise. It's actually more accurate in certain low visibility conditions because it goes through occlusion, but less accurate given certain environments like going under an underpass. Since the noise isn't normally distributed you don't have the guarantees from statistics which you were hoping to rely on.

From the keynote of CVPR. CVPR is the premier annual computer vision event comprising the main conference and several co-located workshops and short courses.

https://cvpr2021.thecvf.com/

https://www.youtube.com/watch?v=g6bOwQdCJrcCVPR

thanks for intro to that car review channel. really good stuff

I don't know much about this but wasn't Tesla getting safety award after safety award in testing and that translating into lower insurance rates?

My understanding is that they are safe in the event of a crash, not that they have the best crash avoidance.

That's patently a hilariously non-sequitur answer to the OP saying HE got rear ended by a Tesla. HE is the one with whiplash and a totaled car. Who CARES about how safe the stupid Tesla was that did it....

So, it is safer "in the event of a crash" for the people inside the car, but if crash rates are higher (statistics pending) it might be more dangerous for everybody both outside and inside of the car.

Exactly. Tesla is 100% trying to keep you safe on the inside while disregarding safety on the outside. Just look at the Cybertruck, pedestrian safety is clearly their last thought.

Doing well in crash tests means the car does well in those specific scenarios, but may not mean much about other safety issues.

The first several model years of the Model S were generally considered to be among the safest vehicles in the history of automobiles.

https://www.businessinsider.com/tesla-tesla-model-s-achieves...

I continue to believe that minor differences car safety ratings are red herrings. (In a previous life I worked as a mech engg in car manufacturing)

It reminds me of benchmarks on MNIST datasets, to decide which model is better. It is like patting yourself on the back for funding better helmets, when your bike infrastructure is straight out of mad max. The real problem is somewhere else.

Crashes almost never occur like we see in these crash tests. The prime culprit is what you can call 'dangerous actions'. If a Tesla driver is far more likely to be distracted (FSD, Touch controls) or more likely to be out of control at speed (insane acceleration with terrible braking), then a Tesla is an unsafe car. This is irrespective of the marginally lower likelihood of death in ideally controlled circumstances. Taking standard customer behaviors into account is one of the basics of design. If you market your lvl2 system as FSD, then a few idiots will let the car drive while in the back seat. If your car can't brake effectively past 120 mph, then maybe it shouldn't accelerate to that number within the blink of an eye.

I am also ignoring the safety risk posed by these vehicles towards anyone else on the street. It has long been my complaint that the SUV & Pickup craze is actively making the roads unsafe. They have worse blind spots & can kill with the kind of efficiency that sedans can only dream of.

Tesla's safety ratings are misleading too. For one, Tesla's high safety has to do with being one of the first truly electric cars on the market. Electric cars can't roll, are heavy and thus the crash zones can absorb more energy due to having more mass. Electric cars are thus objectively safer in straight line collisons. [1] But, that's like saying that your military tank is incredibly safe, but crushing dozens of cars in front of you. Safety should be about how safe the streets are with your car on the road, vs not being on the road. Alas, this is a problem with the car industry at large. So, Tesla isn't really to blame for this one.

[1] https://www.iihs.org/news/detail/with-more-electric-vehicles...

"But, that's like saying that your military tank is incredibly safe, but crushing dozens of cars in front of you. Safety should be about how safe the streets are with your car on the road, vs not being on the road. "

I recently witnessed an almost head-on collision between a H2 Hummer and a Chevy Malibu. The H2 Hummer lost control (i think due to speeding) and served into oncoming traffic. The H2 flipped on it side but was barely damaged; the Malibu was completely mangled.

An "autopilot" that requires a driver's attention is not an autopilot.

If people really understood that the car driving is less reliable than themselves, they might not turn it on at all. Imagine you teaching a 16 year old to drive while commuting, everyday. NTY

> An "autopilot" that requires a driver's attention is not an autopilot.

Weird; I thought that's exactly what an autopilot was.

And they probably still are, along with the rest of the models.

Assuming you're actually the one driving them that is, and not a broken ass camera.

Where I live Tesla has quite high insurance rates. Mostly because it’s very expensive to repair but also because Tesla drivers tend to speed and cause more accidents than other drivers.

Perhaps the drivers didn't mean to speed...

Tesla doesnt make a $200K car.......

They may not be in production yet, but they are asking $200k for the Roadster (and $250k for the founders), which might be the super car I presume parent is talking about. The plaid while not $200k still clocks in at $150k

>They may not be in production yet, but they are asking $200k for the Roadster (and $250k for the founders), which might be the super car I presume parent is talking about

Well, that's the problem. Claiming that the car that isn't even in production yet has "underbraking" issue is misleading at best.

Regardless, I don't think that person was talking about the upcoming Roadster, because they added a link as a citation in "having their $200k super car be criminally underbraked. [2]", and [2] is a link to a video review of Model S Plaid by a big car youtube channel. I have no idea why they linked that specific video, because it was a glowing review of the car. The reviewers were giddy and threw tons of praise at it in pretty much every aspect.

I meant the plaid. Sorry.

You are right, Tesla makes good cars. The rub is pretty much about carelessness on part of Tesla in safety design.

I specifically linked the video because they are even handed in their reviews and not one of those who put down cars just because they don't like the CEO.

Thanks for explaining your point, I really appreciate it.

Mostly because the original reply seemed like it was just trying to insert a video that doesn't fit the criticism drawn in the comment, but is baiting for people who will just read the comment without watching the video. But I get the point you were trying to make now.

While I disagree, I cannot in good conscience not upvote a point made well.