Live data from Hacker News

A surprisingly simple way to foil car thieves

news.umich.edu

421–430 of 576 posts

Re: A surprisingly simple way to foil car thieves

#421
post #169

Earlier quoted context omitted.

No, really, good for them. But don’t be talking down on my car. That’s just not cool. That beast is the workhorse of the farm and it gets the job done. None of the windows roll down though and it’s hot as hell inside, so it discourages unnecessary use, saving the planet. It rarely sees pavement but it drags what needs to be dragged and it pulls the utility rigs out after they deliver to us.

And that's all well and good, but maybe introspect here for a second? You're upset that your accomplishments aren't being respected, immediately after discounting the accomplishments of others. The point is that your car's modifications and the university's are similar, but different, particularly in scale and broad robustness, which adds difficulty in ways you may not be appreciating. $1.2 million may sound like a l…

Did you see the picture of what they built? I wouldn’t describe it as refined or particularly professional.

It sounds like his system is more refined than the academic one. It certainly has more features.

Re: A surprisingly simple way to foil car thieves

#422
post #101
post #16

>At the end of the three-year project, the team aims to have a commercially viable prototype that can be scaled up to commercial production, first as a theft deterrent device, and potentially later as a complete vehicle entry and control system that could replace traditional keys and fobs. What's the point of this? Modern cars already have engines that are cryptographically tied to keys[1]. They're not perfect, but i…

So let me get this straight Modern cars have engines that are cryptographically ties to keys But the Pentagon couldn’t put biometric locks on their humvees? https://www.ibtimes.co.in/isis-takes-dozens-captured-us-humv... They can’t account for trillions of dollars… but we vote them more money they didn’t even ask for https://www.nytimes.com/2019/03/11/us/politics/trump-budget.... ”Mr. Trump’s budget, the largest in f…

Even if you put this device in a car or a biometric device in a car or even a normal keyed system. An enemy who has possession of that car for more than a few hours can easily bypass most locking devices.

This one is nothing more than a relay on the battery line. Simply find the relay and bridge it. Problem solved. Might take you a few hours to dig under the dash to find the damn thing, but once you do 'problem fucking solved'.

Re: A surprisingly simple way to foil car thieves

#423
post #230

Earlier quoted context omitted.

And that's all well and good, but maybe introspect here for a second? You're upset that your accomplishments aren't being respected, immediately after discounting the accomplishments of others. The point is that your car's modifications and the university's are similar, but different, particularly in scale and broad robustness, which adds difficulty in ways you may not be appreciating. $1.2 million may sound like a l…

The professor did gloss over briefly the difficulty in making the system work for a large number of vehicles, before arriving at a viable "signature" idea, as the article describes. Sounds like an area with a lot of false starts (heh) and time consumption, and dead ends.

The professor should have seen that he could send a signature over the airwaves to his relay since that is even more universally compatible… plus, you can buy that exact device for about $20 at the online retailer of your choice.

Re: A surprisingly simple way to foil car thieves

#424
post #153
post #84

1.2 million grant for this? Now I’ve seen grift, but come on. I want to hire their grant writer. Literally a relay in the starter lead. This looks like one of my afternoon projects, and I’m not even joking. I have a 1990s montero diesel and it leaks power, and I often forget to disconnect the terminal. So I bought a relay from AliExpress (same one shown in this photo but one size up) and hooked it up with an esp32 an…

This is very close to what the original LoJack did in 1979. They added keypad to replace switching on lights: [LoJack] could also include the incorporation of a scheme whereby an additional step was required to activate the ignition. Prior to starting, it would require the activation of any number of the usual vehicle features such as the radio, headlight switch, or other switched device. Without knowledge of the pro…

> Modern transponder key based systems made the original LoJack starting system obsolete

Transponder keys aka immobilizer systems.

All those Kias being stolen in the US are being stolen because the US does not mandate any form of immobilizer, and thus Kia on their cheapest models didn't include one.

Canada mandates immobilizer systems. Guess where the whole "Kia boys" phenomenon isn't a problem?

This is one of many examples of how our "democracy" isn't working. The vast majority of the US populace would agree that an immobilizer system which prevents a car from being started with a screwdriver is a good thing.

Every time it's been proposed in congress, the automotive lobby has told congress how very expensive it would be for them (and by very expensive, we're talking probably less than $100 per car.)

The expense to society (the owner losing their likely sole means of transport to work, health care, social activities and suddenly having a massive expense), police response to do something (er, just collect the report, I guess), the lost productivity, emergency services, and medical costs of people injured (victims or perps) from joyriders...all that goes unmentioned, because nobody's spending money to put someone in front of Tommy Tubletone from chucklesville to tell him that it'll cost everyone less to mandate the things.

Consider that ABS was not mandated in the US until 2012, along with traction control.

Compared to a lot of european "socialist" countries, we have much worse alignment between public opinion and legislation, and it's because of how powerful lobbying and corporate election funding is here, and a pervasive, insidious effort to portray anything other than wild-west free-market attitudes as "communism."

Re: A surprisingly simple way to foil car thieves

#426
post #84

1.2 million grant for this? Now I’ve seen grift, but come on. I want to hire their grant writer. Literally a relay in the starter lead. This looks like one of my afternoon projects, and I’m not even joking. I have a 1990s montero diesel and it leaks power, and I often forget to disconnect the terminal. So I bought a relay from AliExpress (same one shown in this photo but one size up) and hooked it up with an esp32 an…

> That way I can basically forget that it exists, problem solved. It has been working flawlessly for two years now. The whole thing took me about 3 hours to put on strip board and program, another half hour to enclose and mount it. Give me an hour more in micropython and I could make it require a passcode entered on your phone with a secret wiper switch sequence as a backup. So, basically a custom rig, with some shit…

No offense to the GP, but geez that comment belongs in some kind of Hacker News hall of fame

Re: A surprisingly simple way to foil car thieves

#427

Earlier quoted context omitted.

Sure. In theory, investors can behave as rational actors. In practice, you get all sorts of chasing the Next Big Thing because there’s too much money flowing around to too little fundamental research. So I’m ok if we spend $1.2M on research. That’s like 5 developers for one year who would otherwise be allocated by the invisible hand into effectively useless endeavors like high frequency trading or ad tech. Cut to the…

> So I’m ok if we spend $1.2M on research. I'm not. Where do I opt out of funding government grants? At least I can decide not invest in something if I don't want to risk losing money.

Government funded research gave us computers, the internet, EUV technology, GPS, lasers, and all sorts of other shit.

If you're on HN in the first place, there is an approximately zero percent chance that the few dollars of your tax money spent on this kind of research hold a candle to the benefits you've gained from it.

Re: A surprisingly simple way to foil car thieves

#428

I knew someone with a shitty pickup truck where he wired a light switch into the fuel pump and hid it under the steering column. The truck would be stolen something like 5 times a year and every time he would find it abandoned a km or so away when the gas in the line ran out. I drove in it once to get lunch and near the beginning the car begins sputtering and he's like, "oh right, you're thirsty" and reached under an…

My last century GTI had kill switch (it was one of those circular keys and a spring loaded button) under the steering wheel. The engine would turn like it was trying to start, but it wouldn't unless it was unlocked. The other thing the switch did which I didn't realize at first was lock the hood. I bough the car used and trying to lift the hood initially I thought the latch was broken..

Re: A surprisingly simple way to foil car thieves

#429
post #413

Earlier quoted context omitted.

This is a frustrating yet common sort of take. Yes, this is simple, as the article clearly points out. Yes it is obvious in retrospect. But did you do anything with your brilliant work besides bodge your terrible car a few more miles down the road? There is value to developing the entire system... to ensuring the keypad mechanism is reasonably robust and tamper proof. There is value to understanding the vehicle as a…

There is zero novel research here, and the entire purpose of the 1.2 million dollar grant was research. All the value you are mentioning is related to bringing a product to market, which is something that the grant did not require and universities don't usually follow through. Most of the time transition to industry happens is when there are motivated companies who do all the work to bring the device to market, but n…

I disagree: the device monitors battery fluctuations to 'authenticate' the driver. the fluctuations need to be a specific pattern - delivered either by a device plugged into the 12v accessory port, or by some specific pattern of driver behavior, such as quickly flashing lights, activating wipers, etc. This is indeed a novel approach.

And it is a fair sight more involved than a simple kill switch, by the look of things. The research aspect comes from exploring the practicality of such an approach. This exploration requires prototypes, test beds and investigators.

Who's really to say what the results of the research will be, at this point? In my opinion, I think smart phones and NFC are probably the way to go... but I'm not going to hop on the internet and make scornful remarks until I know more. I'm not sure why you have done so?

Re: A surprisingly simple way to foil car thieves

#430
post #22
post #9

Earlier quoted context omitted.

I have an old beater. It cost me 2300 USD when I bought it. Last year the ignition cylinder broke. My friend is a mechanic and he told me he could replace the cylinder for 90 USD, or buy him a beer to just install a button. I now jokingly call my car a Tesla since it starts with a button (it also requires the key)

Tesla cars don't need a button to start

The brake pedal is _almost_ like a button.
Post reply on HN