Connected cars aren’t just for hacking

We visited the Connected Car Expo and found some interesting new applications.

Read the whole story

 

[Nowicki]

Having cars drive them selfs is an amazing thing so long as it doesnt need to communicate with anything to actually drive. The second you let something communicate with a car thats driving itself it can and probably will have security flaws that could cause some serious accidents. I know companies want to be able to know if your driving by a billboard so that their analytics are complete, but this is not worth it from the consumers perspective

So if the car reads its environment from lasers, radar like devices, and cameras im ok with that.

 

[potato44819]

"Love it or hate it—and we know many of you hate it—the connected car isn't going away."

We don't hate it. We want it to be implemented with safeguards that we mere humans can monitor.

 

[Shuffy]

The only thing I require is that the car can function perfectly fine as a car without connecting to the outside world. If it becomes a requirement to function, then I feel we're screwed.

 

[Drizzt321]

That Voyomotive sounds interesting, and while 256-bit keys and independent security reviews _sound_ good, I'd be very curious to see their reaction by other security researchers that might find something and how that all will get handled. And if there are OTA updates to the device, and how those are handled.

 

[Whiner42]

Modern Major General Thanatos[/url]":2hjx4qdc]"Love it or hate it—and we know many of you hate it—the connected car isn't going away."

We don't hate it. We want it to be implemented with safeguards that we mere humans can monitor.

And we don't want our cars spying on us without our permission and sharing the data with ... who knows? The manufacturer? Advertisers? Law enforcement? Insurance companies?

 

[giltwist]

I just bought a new Chevy. First thing I did was completely remove the onstar module. The in-dash compass stopped working, but that's easily replaced with a stand-alone GPS. Both terrestrial and satellite radio work. The blue-tooth is a bit wonky; but I'm hoping that, with some trial and error, I can reconnect the bluetooth antenna without reconnecting the cellular antenna. Even if I can't get the bluetooth working, that's a VERY small price to pay to know my car is secure from wireless hacking.

 

[Drizzt321]

giltwist[/url]":1rtxgqb4]I just bought a new Chevy. First thing I did was completely remove the onstar module. The in-dash compass stopped working, but that's easily replaced with a stand-alone GPS. Both terrestrial and satellite radio work. The blue-tooth is a bit wonky; but I'm hoping that, with some trial and error, I can reconnect the bluetooth antenna without reconnecting the cellular antenna. Even if I can't get the bluetooth working, that's a VERY small price to pay to know my car is secure from wireless hacking.

Wait, you can totally remove that module? Is there a way to get the vehicle without that module installed?

 

[Old_Fogie_Late_Bloomer]

[T]here aren't really any bad cars.

As a former TDI owner, I have to say I have trouble taking this statement seriously. I liked my Passat--it was both comfy and torque-y--but I'm glad I don't have to be seen driving it.

 

[jdale]

"We have an independent security review being conducted by a major research lab... they're going to have full access to how it's implemented, other than the keys," he told us.

Really, this should be required for automobile connectivity, just like independent testing of emissions (hmm) and crash safety. It's a safety issue and as such should not be left to the manufacturer to certify itself as good enough.

But good for them for doing it.

 

[svgpithon]

At some point I want to do a DIY car infotainment system. There was a Danish(?) member of the Bit-Tech forums who had a custom liquid-cooled rig fitted in his, with custom dash and everything. That was neat.

 

[Chuunen Baka]

I try to make a new car last 10 years or so. I don't think I have any tech or software that old. Anything shipped with a new vehicle is going to deprecate faster than its host.

 

[Tazmjm69]

"Eventually our cars and roads will all be equipped with vehicle-to-vehicle and vehicle-to-infrastructure systems, but widespread deployment of V2V and V2I is at least two decades away. Harnessing the power of our phones to achieve similar goals in the meantime seems like a no-brainer."

At least 2 decades away? With the rate that technology is expanding, you seriously think this is an accurate statement?

 

[Dr Gitlin]

Tazmjm69[/url]"45zvpoi]"Eventually our cars and roads will all be equipped with vehicle-to-vehicle and vehicle-to-infrastructure systems, but widespread deployment of V2V and V2I is at least two decades away. Harnessing the power of our phones to achieve similar goals in the meantime seems like a no-brainer."

At least 2 decades away? With the rate that technology is expanding, you seriously think this is an accurate statement?

Before it's ubiquitious? Sadly, yes. People are keeping hold of their cars for longer and longer (now 11 years on average), and the states and municipalities barely have money to pay for basic road repairs, let alone fancy 802.11 streetlights.

 

[Eraserhead]

Are manufacturers going to guarantee 20 years of software updates for smart cars?

 

[Leachpunk]

Chuunen Baka[/url]":1ab42bv6]I try to make a new car last 10 years or so. I don't think I have any tech or software that old. Anything shipped with a new vehicle is going to deprecate faster than its host.

I was thinking about this myself. It's going to reach a point that consumers will be buying cars based on whether the tech inside is upgrade-able or not. Even then, with how swiftly technology changes (not looking at ATX) the whole motherboard and corresponding components may need to be completely swapped out. Basically I imagine this can only really be achieved if most cars have a standard mounting area for the on board computer system.

 

[xme]

I hadn't seen the Voyo before today, but I really prefer solutions like that over fully integrated smart-things. Kind of like my TV and Roku/ATV/whatever.. Keep the large, expensive part dumb, use standard interconnects, and replace the outdated bits as needed. I'm too cynical to believe this is the direction things will go though.

 

[daarong]

Reliability and safety have made such strides that car makers are looking at digital bells and whistles as a way to differentiate their products from competitors

How about a decent warranty on the electronics? The manufacturers are getting comfortable with the mechanics, so we're seeing ever longer mechanical warranties... but it doesn't appear the electronics are in the same boat. And the electronics are NOT cheap to repair or replace.

 

[giltwist]

Drizzt321[/url]":3egrsd7g]

giltwist[/url]":3egrsd7g]I just bought a new Chevy. First thing I did was completely remove the onstar module. The in-dash compass stopped working, but that's easily replaced with a stand-alone GPS. Both terrestrial and satellite radio work. The blue-tooth is a bit wonky; but I'm hoping that, with some trial and error, I can reconnect the bluetooth antenna without reconnecting the cellular antenna. Even if I can't get the bluetooth working, that's a VERY small price to pay to know my car is secure from wireless hacking.

Wait, you can totally remove that module? Is there a way to get the vehicle without that module installed?

In many vehicles, the onstar module is just a discrete box about the size of a hardback book. In my particular car, it was behind the glove compartment, but I've seen it in a trunk side-panel as well as under the dash. Based upon my googling, some vehicles fare better or worse when removing the module entirely, but the procedure is basically just a matter of unplugging it.

I removed mine entirely, but got the symptoms mentioned in my earlier post. This video directed me to try opening the box. The internals in mine weren't identical to what was in the video, but close enough that I think I may have removed the board with the cellular antenna. I haven't tried reconnecting it yet, though.

 

[SGJ]

It's not just the security of the Voyo device that needs to be considered.

I wouldn't be happy with a device that sends commands through the OBDII port based on "reverse engineering" a particular manufacturer's implementation. Manufacturer's occasionally get things wrong and they don't have to reverse engineer (i.e. guess) anything. Who is going to take responsibility when something goes catastrophically wrong? The car manufacturer, Voyomotive or the driver who fitted it to their car?

Safety critical systems shouldn't be controlled by add-on software developed without full details of the protocols and systems being controlled and a full safety case being developed.

 

[ranthog]

Dr Gitlin[/url]":1qz0kccx]

Tazmjm69[/url]":1qz0kccx]"Eventually our cars and roads will all be equipped with vehicle-to-vehicle and vehicle-to-infrastructure systems, but widespread deployment of V2V and V2I is at least two decades away. Harnessing the power of our phones to achieve similar goals in the meantime seems like a no-brainer."

At least 2 decades away? With the rate that technology is expanding, you seriously think this is an accurate statement?

Before it's ubiquitious? Sadly, yes. People are keeping hold of their cars for longer and longer (now 11 years on average), and the states and municipalities barely have money to pay for basic road repairs, let alone fancy 802.11 streetlights.

Well even without V2I, V2V has some great benefits. Unfortunately, some V2I needs to exist for certificate revocation.

The research on both the deployment phases, and what happens during natural disasters is interesting. Part of the good news is that even if deployment is slow, it doesn't take very many strategically placed V2I nodes for the messages to be quickly passed around a city thanks to V2V message passing.

There is some V2I benefits you won't get in a sparsely populated network, but at the very least it could still support most of the important things that V2V brings us.