[Cynystyr]

I don't get the point of this discussion at all...
Google developed Android Auto. It is their system.
This new system that GM is using is also developed by Google...
If you like Android Auto, then there is no reason you shouldn't give the new system an honest chance. Google should know their stuff pretty darn well, and this new system could implement changes that are improvements over the anitquated original Google Auto ... err Android Auto.

 

[ontheroad]

Check out the new Kia EV3 that's coming out at the end of the month. Based on Kia build quality it may turn out to be a worthy contender.

 

[Steverino]

I don't get the point of this discussion at all...
Google developed Android Auto. It is their system.
This new system that GM is using is also developed by Google...
If you like Android Auto, then there is no reason you shouldn't give the new system an honest chance.

 

[hellsop]

I like the idea of built-in Google Maps as it will be up to date with all the charging stations and will calculate the route to the charging station base on more data like weather, temperature, state of charge and elevations.

... and make recommendations about what stations need to be stopped at based on availability, capacity, and rate of consumption currently driving, and figure how long stops need to be to make the next step, etc.

 

[hellsop]

I like android auto. When I am taking one of my kids' friends home, i simply hand them my phone, usually in the backseat, and say enter your address, and boom, I am navigating. The current system is fine and works great. The new system is a money grab, plain and simple.

Is it enough to keep me from buying an Equinox EV? I'm not sure yet, but its a pretty big deterrent for me. I don't want to just use a phone holder and google maps, the proper screen looks much better IMO.

And you won't be able to do a similar thing with the in-dash system? "Hey kid, what's your address?" "Okay google, drive to 123 First Street, Mapleton."

(If the Google Assistant in Android Automotive works half as well-integrated as Google Assistant is in Home, it won't even matter whether your phone is listening or the car, because it'll know to put a routing right into the car as soon as the car comes available anyway.)

 

[Andre YUL]

If you like Ultium and are OK with something closer to the Blazer's price and size, checkout the Honda Prologue. It has wireless CarPlay and AA.

That said, if you want optimal battery pre-conditioning and route planning that maximizes the charging capabilities of your EV, you're better off with the built-in navigation.

 

[Dtownboys]

Thinking i could get trapped in costly subscription technologies like mandatory internet plan with onStar with subscription for navigation and supercruise, i think i've made the decision to look for another car (i had made a deposit on this car already at dealership).

What other car i should look into, i know there is the EV6 or the Ioniq... maybe the Mustang... i don't know.

What other cars are you looking into (for those who are not 100% sure going with the Equinox)?

There is a lot of discussion these days about GM’s (and others) decision to move beyond Carplay and Android Auto when it comes to the on-board electronics that will run electric vehicles. I have a perspective on this that is formed by 37 years of work in the field of industrial process control automation.

Some industrial processing plants are quite dangerous. A loss of control over the process results in massive explosions (example: gasoline production). In today’s world, these types of processing plants have been made quite safe through the use of industrial-grade automation. Computer control is simply more reliable than human control.

But I must clarify that industrial-grade automation is performed by purpose-built computers that run 24 x 7 x 365 without error. These computers are not general purpose personal computers. They do not text, send-mail, search the internet and make phone calls. Rather, these computers have one job and that is to run the industrial process. They exhibit what we call “mission critical robustness”.

With this concept in mind, I see a number of direct lessons for car manufacturers. I have taken a keen interest in autonomous driving. I firmly believe that at some point, we will find that computers are better (safer) at driving than humans. Computers do not get drowsy, drunk and distracted. However, autonomous driving computers must be built with mission critical robustness.

How does this relate to CarPlay and Android Auto? These general purpose applications are fine for playing Spotify over your car sound system or managing hands-free phone calls. But, once a car embarks on autonomous driving, we are in the realm of mission critical.

I would never want a smartphone driving my car. There are simply too many unknowns when two very different computers are trying to work together. I want one purpose-built system to control the show. Have you ever seen the CarPlay connection to your car temporarily drop out? Have you ever gotten an update to your Android phone and your mapping software behavior changes? These types of issues can not be tolerated in a mission critical environment. This fails the robustness test.

I’ll further explain that by autonomous driving I am not referring only to the ultimate level 5 self-driving car. Starting with level 2 driving such as provided by GM Supercruise, it is imperative to build on a computer system that is purpose-built. When you move beyond adaptive cruise control to an environment where the hands are not on the wheel and the car is changing lanes without human intervention, I believe that the game becomes serious. And as we take steps towards level 5, we must be gaining run-time experience on a computer platform that has the ultimate objective in mind.

I am assuming then that mission critical robustness is a fundamental reason behind GM’s decision to move away from CarPlay and Android Auto. And I believe this is the right decision for now. Maybe in the future, we can return to a situation where smartphone integration is done in a way that partitions the computing environment to where the phone does only phone things. But for now, the simplicity of a single purpose-built system is what makes sense to me.

 

[drew]

There is a lot of discussion these days about GM’s (and others) decision to move beyond Carplay and Android Auto when it comes to the on-board electronics that will run electric vehicles. I have a perspective on this that is formed by 37 years of work in the field of industrial process control automation.

Some industrial processing plants are quite dangerous. A loss of control over the process results in massive explosions (example: gasoline production). In today’s world, these types of processing plants have been made quite safe through the use of industrial-grade automation. Computer control is simply more reliable than human control.

But I must clarify that industrial-grade automation is performed by purpose-built computers that run 24 x 7 x 365 without error. These computers are not general purpose personal computers. They do not text, send-mail, search the internet and make phone calls. Rather, these computers have one job and that is to run the industrial process. They exhibit what we call “mission critical robustness”.

With this concept in mind, I see a number of direct lessons for car manufacturers. I have taken a keen interest in autonomous driving. I firmly believe that at some point, we will find that computers are better (safer) at driving than humans. Computers do not get drowsy, drunk and distracted. However, autonomous driving computers must be built with mission critical robustness.

How does this relate to CarPlay and Android Auto? These general purpose applications are fine for playing Spotify over your car sound system or managing hands-free phone calls. But, once a car embarks on autonomous driving, we are in the realm of mission critical.

I would never want a smartphone driving my car. There are simply too many unknowns when two very different computers are trying to work together. I want one purpose-built system to control the show. Have you ever seen the CarPlay connection to your car temporarily drop out? Have you ever gotten an update to your Android phone and your mapping software behavior changes? These types of issues can not be tolerated in a mission critical environment. This fails the robustness test.

I’ll further explain that by autonomous driving I am not referring only to the ultimate level 5 self-driving car. Starting with level 2 driving such as provided by GM Supercruise, it is imperative to build on a computer system that is purpose-built. When you move beyond adaptive cruise control to an environment where the hands are not on the wheel and the car is changing lanes without human intervention, I believe that the game becomes serious. And as we take steps towards level 5, we must be gaining run-time experience on a computer platform that has the ultimate objective in mind.

I am assuming then that mission critical robustness is a fundamental reason behind GM’s decision to move away from CarPlay and Android Auto. And I believe this is the right decision for now. Maybe in the future, we can return to a situation where smartphone integration is done in a way that partitions the computing environment to where the phone does only phone things. But for now, the simplicity of a single purpose-built system is what makes sense to me.

I have a very different view on this. I think all critical systems (automation, control, etc) need to be segregated from infotainment systems. Allowing people to download apps into a system that is controlling critical functioning is a problem. Allowing phone mirroring to cover functions that are not related to the operation of the vehicle, such as music apps, keeps them from potentially affecting vital systems. If you want automation to work with nav, using onboard nav makes sense, but I'd definitely want it tightly controlled to avoid potential malicious code.

 

[Andre YUL]

I'm wondering if EV makers that exclusively have built-in navigation such as GM, Tesla and Rivian have a reduced risk of car theft? If day-to-day vehicle navigation, not just occasional OTA updates, require wireless connection from the manufacturer, I'm thinking such vehicles would be less attractive for resale. There could still be theft for car parts, but there would be much less demand for shipping and resale overseas.

 

[obermd]

There is a lot of discussion these days about GM’s (and others) decision to move beyond Carplay and Android Auto when it comes to the on-board electronics that will run electric vehicles. I have a perspective on this that is formed by 37 years of work in the field of industrial process control automation.

Some industrial processing plants are quite dangerous. A loss of control over the process results in massive explosions (example: gasoline production). In today’s world, these types of processing plants have been made quite safe through the use of industrial-grade automation. Computer control is simply more reliable than human control.

But I must clarify that industrial-grade automation is performed by purpose-built computers that run 24 x 7 x 365 without error. These computers are not general purpose personal computers. They do not text, send-mail, search the internet and make phone calls. Rather, these computers have one job and that is to run the industrial process. They exhibit what we call “mission critical robustness”.

With this concept in mind, I see a number of direct lessons for car manufacturers. I have taken a keen interest in autonomous driving. I firmly believe that at some point, we will find that computers are better (safer) at driving than humans. Computers do not get drowsy, drunk and distracted. However, autonomous driving computers must be built with mission critical robustness.

How does this relate to CarPlay and Android Auto? These general purpose applications are fine for playing Spotify over your car sound system or managing hands-free phone calls. But, once a car embarks on autonomous driving, we are in the realm of mission critical.

I would never want a smartphone driving my car. There are simply too many unknowns when two very different computers are trying to work together. I want one purpose-built system to control the show. Have you ever seen the CarPlay connection to your car temporarily drop out? Have you ever gotten an update to your Android phone and your mapping software behavior changes? These types of issues can not be tolerated in a mission critical environment. This fails the robustness test.

I’ll further explain that by autonomous driving I am not referring only to the ultimate level 5 self-driving car. Starting with level 2 driving such as provided by GM Supercruise, it is imperative to build on a computer system that is purpose-built. When you move beyond adaptive cruise control to an environment where the hands are not on the wheel and the car is changing lanes without human intervention, I believe that the game becomes serious. And as we take steps towards level 5, we must be gaining run-time experience on a computer platform that has the ultimate objective in mind.

I am assuming then that mission critical robustness is a fundamental reason behind GM’s decision to move away from CarPlay and Android Auto. And I believe this is the right decision for now. Maybe in the future, we can return to a situation where smartphone integration is done in a way that partitions the computing environment to where the phone does only phone things. But for now, the simplicity of a single purpose-built system is what makes sense to me.

I have a very different view on this. I think all critical systems (automation, control, etc) need to be segregated from infotainment systems. Allowing people to download apps into a system that is controlling critical functioning is a problem. Allowing phone mirroring to cover functions that are not related to the operation of the vehicle, such as music apps, keeps them from potentially affecting vital systems. If you want automation to work with nav, using onboard nav makes sense, but I'd definitely want it tightly controlled to avoid potential malicious code.

As a CIO with a software engineering background, you're both right. I'm concerned GM and others aren't properly segmenting their in-vehicle networks with vehicle operations interfacing with the infotainment system controls in a well defined, heavily debugged and parameter checked interface. To do this automakers will need to actually install a local area network in the vehicles and use industry standard techniques to ensure both the heavily protected vehicle operation and control software and the user interface and creature comfort controls can run on the same set of wiring. Tesla is ahead of everyone else here with their Cybertruck having a local area wired ethernet built into the vehicle. GM's Ultium uses a wireless based ethernet for battery management - I do wonder just how secure it is. Wi-Fi can be secured when implemented properly, but most commercial installations don't support the level of security Ultium/Ultify requires.

Also, long term, Android Automotive will need to be replaced for the vehicle control software. It's not a RTOS (Real Time Operating System), and this is what's required for all mission critical embedded systems. Linux based systems can be converted to RTOS; SpaceX has done just this for Falcon 9, Cargo Dragon, Crew Dragon, and Starship, but Android is not a RTOS by design.

 

[Steverino]

GM's Ultium uses a wireless based ethernet for battery management - I do wonder just how secure it is. Wi-Fi can be secured when implemented properly

Visteon is the wireless BMS maker GM partnered with for the Ultium battery platform. They use the word "security" as a bullet point but don't going any real detail about what that means.

I did find:

"The company has been involved in BMS development since the earliest days of EV production when it supported the 1998 Ford Ranger EV. The wireless BMS uses chips from Analog Devices to eliminate the traditional wired harness, saving up to 90% of the wiring and up to 15% of the volume in the battery pack, as well as improving design flexibility and manufacturability.

The chips include all the integrated circuits, hardware and software for power, battery management, RF comms and system functions in a single, system-level product that supports ASIL-D safety and module-level security."

 

[Dtownboys]

I have a very different view on this. I think all critical systems (automation, control, etc) need to be segregated from infotainment systems. Allowing people to download apps into a system that is controlling critical functioning is a problem. Allowing phone mirroring to cover functions that are not related to the operation of the vehicle, such as music apps, keeps them from potentially affecting vital systems. If you want automation to work with nav, using onboard nav makes sense, but I'd definitely want it tightly controlled to avoid potential malicious code.

 

[Dtownboys]

I agree with what is being written here; good thoughts. I think the most basic thing I like about GM's Google Builtin architecture is that the maps aren't being served from a phone. I hope to soon be using Supercruise to drive my car and I certainly don't want my phone involved. Just last night I was driving my Lexus hybrid with Android Auto providing the navigation while listening to Amazon Music, also on my phone. The map dropped out for about 15 seconds and I was not given a turn instruction. My guess is that the phone's processor got overloaded and it prioritized The Allman Brothers over the right turn that was coming up. Maybe this would be ok with Level 2 automation (like the current version of Supercruise) because I must be ready to take over at all times. But when we are at Level 3 automation and I don't need to be paying attention....that seems really bad.

The point that Drew made seems logical, that you'd want the phone to take care of phone things while the car takes care of driving things makes sense. But I worry that bullet-proofing the driving functions from being compromised by the phone functions is a hard problem. Back to my process plant automation background, we found it very difficult to build this partitioning and have it work with 99.9999% robustness. I think it can be done but it will take time.

 

[GreatLakesEqEV]

Thinking i could get trapped in costly subscription technologies like mandatory internet plan with onStar with subscription for navigation and supercruise, i think i've made the decision to look for another car (i had made a deposit on this car already at dealership).

What other car i should look into, i know there is the EV6 or the Ioniq... maybe the Mustang... i don't know.

What other cars are you looking into (for those who are not 100% sure going with the Equinox)?

Recently leased an Eq EV, the RS trim. Super Cruise and App access (built-in Android on the car) are included for 3 years. OnStar was optional.

It seems like these things are negotiable at the time of purchase/lease. Ask the dealer to through hem in or extend. It's software so the manufacturer isn't losing much including it.

 

[Johnge Fredrick]

Thinking i could get trapped in costly subscription technologies like mandatory internet plan with onStar with subscription for navigation and supercruise, i think i've made the decision to look for another car (i had made a deposit on this car already at dealership).

What other car i should look into, i know there is the EV6 or the Ioniq... maybe the Mustang... i don't know.

What other cars are you looking into (for those who are not 100% sure going with the Equinox)?

The same here, we opted for Toyota because of no Android Auto. We know from past experience what a rip off Onstar was

 

[drm]

The same here, we opted for Toyota because of no Android Auto. We know from past experience what a rip off Onstar was

Onstar is not mandatory, many of us do not get it at all. If you are going to keep the car for many years, it would become an issue.

 

[BehindBars]

The same here, we opted for Toyota because of no Android Auto. We know from past experience what a rip off Onstar was

I don’t consider free Onstar for 8 years in my Equinox EV a ripoff. I don’t even have Onstar in my Bolt because I saw no need, so no ripoff there, either. What Toyota EV did you buy?

 

[Andre YUL]

I don’t consider free Onstar for 8 years in my Equinox EV a ripoff. I don’t even have Onstar in my Bolt because I saw no need, so no ripoff there, either. What Toyota EV did you buy?

There is only one Toyota EV in America, the bZ4x. It has even worse charging than the Equinox.

Its depreciation so high (higher than even other EVs), that you could afford to pay for many years of OnStar with the higher resale value of of a GM EV. I went from Android Auto to the built-in Google Automotive. I'm fine with the navigation and connectivity that are included for 8 years.

If, for reasons that don't resonnate with me, you feel you absolutely need CarPlay or Android Automotive, rather than the Solterra/bZ4x compliance cars, I'd suggest looking at the Honda Prologue, one of the very nice Hyundai or Kia EVs, the Mustang Mach-E or maybe even a Nissan Ariya. On a lease, I might even prefer a Vinfast to a bZ4x!

Only reason to go Toyota over the preceeding, IMO, is if you really need the ground clearance. Then again, if that's your motivation, the Solterra is slightly better.

 

[drm]

If you were looking at the cheaper LT Equinox trim, then there really is not another EV at that price with similar range. Of course if range is not an issue because you just want an around town errand car, then a used Bolt or a Kona EV could be the thing.

If you were considering the more expensive Eq trims, then the Hyundais and Kias are getting close in price, but their base models will still not match the Equinox range.

 

[Andre YUL]

If you do frequent road trips and have good charging infrastructure in your area, then some of the Hyundai & Kia offer impressive charging speeds, efficient use of space and quite nice interior materials. The style of the Ioniq 5 and Ioniq 6 is more polarizing.

In my area, charging infrastructure that would make an Ioniq 6 significantly better charging than the Equinox EV is rare. We're planning a trip in Ontario (Niagara peninsula on Lake Erie) where we'd level 1 charge at the rental and have to drive 15 minutes for the nearest 50kW charger... We may take my wife's ICE car.