• 9 posts
  • 33 comments
Joined 3 years ago
Cake day: July 8th, 2023
  • I had GPSD working nicely, confirmed both by cgps -s and mongps, but I was stuck on geoclue. Whatever I did, the GPS wouldn’t send data to Organic Maps. I kept having this error: “Error: GDBus.Error:org.freedesktop.DBus.Error.UnknownMethod: Object does not exist at path “/org/freedesktop/GeoClue2/Client/1”” despite being 100% sure client1 was the right one.

    Going to Lineage OS was much easier for me and I just needed a working GPS. Linux GPS will stay in the “challenge todo list”, because I’d like to figure it out eventually. I most likely was doing something stupid.

  • What car do you have? Dacia Sandero Stepway

    Didn’t you want to use it for your music as well with navidrome? Maybe, if the USB port works, then I’ll use Navidrome on it and put the music that way through the car speaker. If they don’t, I can plug my phone to the car speakers and the Pi to bluetooth external speakers. The phone only crashes during satnav. By the way I did a recovery reset, i’ll try a drive today and see how it goes.

    CoMaps may be up your alley, uses OSM. Thanks for the name, current app in testing is OsmAnd, as i’ve seen it recommended a lot. If it doesn’t work I’ll try CoMaps.

    Especially since I just bought the car a few weeks before. I feel you, this phone was bought specifically for car use. My former one was slow, but I don’t use my phone much so there was no point in upgrading. I got the Oppo A51 back then because it had decent reviews for the price and I had a deal on it. But I think it’s not beefy enough. It thermal throttles very easily, too.

  • I didn’t know about osmin! I might try it out.

    For the poweroff, I’ll just poweroff the Pi from the UI and wait a sec for it to go down. Initial testing has me waiting 10 seconds on the “shutting down” screen. That’s short enough, meanwhile I can put my jacket on or something. Once off, I just unplug it. It’s also running on SSD because I don’t trust SD cards to endure constant read/write from satnav so I installed from SD card and cloned the OS to an SSD, it should be more resilient this way.

  • For the audio, I’ll first try the obvious and plug the Pi to the USB A of the factory car unit. Maybe it’ll detect it as mass storage media and give it access to the built in audio system. Else, I’ll just use a bluetooth speaker I have at home. I don’t need good audio for “turn left/turn right”, just to hear it.

    The whole project is an attempt at getting rid of Android Auto and having my own standalone unit that will free me from unwanted updates both from the OS and the apps. I’ve had Waze become useless three times in the span of two years because they pushed updates that made the app unstable enough to not be reliable.

    For the viewing angle, on the car next to the existing head unit screen, there’s a mounting bracket. I’ll use that to mount a 7inch display I got. It’ll sit right above the existing one if that doesn’t obstruct the view too much, else I’ll have to get one of those arms with joints and lower it somewhere not annoying.

    My bluetooth speaker is the perfect size to sit in the cup holder so if the car audio isn’t possible, then there’s a plan B.

    I’ll try the recovery mode for my phone and see if that helps!

  • Hey, thanks for the link. I haven’t looked at that type of parts because I’m not confident tinkering with the car, especially since it’s still fairly new and I’m too afraid to damage some plastic. And I also don’"t have any spare room available on the dashboard anywhere.

    There’s a little nook to empty your pocket between the gear shifter (manual car) and the dashboard but it’s wide open, fairly hot, and probably still too small even to fit a raspberry pi in a case with the cooling I’ll need.

    That’s why I’ve been looking at external solutions. If wires weren’t an issue, I’d put a trunk/boot organizer at the back and store the pi unit there during drives but it’s not practical with the wiring. Still routing cables in my head in case I’d get an idea.

    Here’s a photo I found of the front of the car: https://www.ouestfrance-auto.com/sites/default/files/sandero_stepway_1_0.jpg

Hello everyone!

I already posted a first time here a while ago: https://lemmy.world/post/30549957

My goal was to replace my Android Auto + phone setup for satnav in car with a dedicated Raspberry Pi.

Here’s a status update of the project so far!

Choosing the parts and getting them here was obviously fairly easy even though it took quite bit of research to find a way to power the Pi4 in car reliably.

It’s a pi4 with an adafruit ultimate gps dongle, an SD card and a 2.5 SSD. The SD for OS installs, and then I clone them to SSD for better stability and performances.

I first tried to go Pi OS route. I figured Linux might open more possibilities and I’m more comfortable with Linux. I tried a few options in Pi OS, namely Navit. I banged my head quite hard on that one, trying to figure out how to make it work, but I never managed to get a good navit.xml config file. After hours upon hours of trials, I gave up.

Tried Organic Maps then, but it was a flatpak which introduced a lot of permission issues and I never got the GPS dongle to talk to the app.

After a few failed attempts, I decided to try Emteria OS, an android spin available in the Pi Imager. It doesn’t boot without SD card as opposed to PiOS, so I was glad I spent the 5€ for the SD card. It booted easily, SSH was harder to enable than on Linux but it’s likely because I just know Linux better. Installing apps was easy, but I stopped there and didn’t even try to get the GPS working because 90% of my RAM was used idling at boot, which makes the whole project impossible on Emteria. Not sure if there’s a bug in the current version or what, but I simply moved on.

Then I went to LineageOS. Similarly to Emteria, installing apps was very easy. Getting the GPS from Adafruit to talk to Android was fairly easy if you read the docs carefully. I needed to install android dev tools on my main PC and connect to the Pi using ADB as root to edit some config files. That’s it, the GPS worked and I had a working GPS unit.

Now I’m at the point where I need to introduce the package into the car. For now , the pi still doesn’t have a case because I didn’t want to limit my options in the car. I still haven’t found a good way to bring the pi with me and have it being safely transported without breaking or even becoming a deadly cannonball in case of crash.

My current idea that I’m exploring and checking is getting a Pelican Case, not sure if any of their cases has vent holes. The pi would go inside the case, and the case would be attached to a strong anchor point like the seat rails. Not sure how, yet. I’m thinking maybe about carabiners from Petzl since they would be much stronger than needed, I’m just not sure the anchor point on the case would be strong enough.

So there you have it, I’ve made good progress on the project and I’m confident it will work out reasonably well.

  • That’s the obvious route, indeed. But it’s not a fun one, I don’t get to own the data and the device is subject to poor updates which will render it useless in a few years. I’d rather buy an old GPS than buy a new phone honestly. The phone that is currently the culprit cost me 230, 23 months ago. It still works perfectly fine outside of the car for everything I need it for, it just doesn’t like being used in car. It’s too much for it I guess, though it wasn’t too much two years ago.

    I considered a tablet, but I’m not sure if there’s a serious advantage over a phone at that point.

  • I didn’t take it as throwing shade :) I just thought I’d provide more context, though reading it back now it looks more like a rant.

    Another person here pointed me to the Crankshaft too. It looks cool but it still uses a phone. I’ll try to erase this silicon slab from the equation entirely. From my personal experience over the years, phones have become a hindrance more than a supportive gadget.

    I think i’ll go ahead and give it a try. I just need to figure out a way to cool the device properly in a hot environment.

  • Yeah, possibly, I might be a contrarian but I’m tired of cloud services, subscriptions, big corporations. I’m also tired of multipurpose devices that do plenty of things poorly. I just want a screen that shows me how to get somewhere, reliably. My phone from two years ago simply stopped being recognized by the car, bought a new one for that single purpose, and this fella already crashes too.

    I didn’t go crazy on the budget but it worked initially and now it doesn’t at all so I’m tired of this bad tech. If I build my own, my hope is that I can make it more resilient and easier to repair/fix.

  • That’s the thing, the phone never crashes otherwise. I don’t really use it anyway because I have zero use and interest for phones, it’s a device I have for 2FA, emergencies and satnav. My current one I bought 20 months ago at 230€. Fairly cheap, but not the cheapest either. In that short span of time, Waze had a total of 3 bad updates that forces me to go back to Google Maps until they patched it.

    The whole reason behind the project is to get rid of the phone factor because it simply is unreliable. If not for the fun side of the project, I’d simply get an old school GPS that goes on the cigarette lighter.

    I thought of a Pi because that’s the established brand, but if something else works just fine, then I might just go for it.

    From some troubleshooting I made, apparently my phone heats up in the car because it automatically charges. The lower the battery, the harder it wants to charge and the more heats it generates. However, it doesn’t shed excess it fast enough so there’s a buildup until it crashes.

    My car is, indeed, pretty hot. Getting a built device might solve it, or not, or temporarily until the device gets older, depending on the thermal management. Making my own device lets me handle it the way I see fit and go nuts on cooling if I want to.

    I’ll take a look at the OSM screen update settings, that might actually help!

  • I have zero experience tinkering with cars so my first draft for power was simply a huge powerbank or two. In my use case, the device is meant for use when I drive alone, and when that’s the case it’s mostly very short trips of maximum 1 hour with standstill traffic, but more often than not max 30 minutes. Down the line I might improve the system and wire things more permanently but I need to build confidence first.

    I was also looking at using the cigar plug for power, potentially, but I like the idea of a self sufficient, removable portable device.

Hello everyone,

I’m thinking about a project and would like to ask for a second opinion from more experienced people. I sadly didn’t find a community dedicated to that on Lemmy and here’s the closest I know about. Let me know if I need to move the subject elsewhere, I understand this is on the fringe.

I have experience self-hosting many things on an old gaming PC at home.

Recently my phone which I use for music (navidrome) and satnav in car via Android Auto keeps crashing. The easiest solution would be to get a new phone but this one isn’t even two years old so I’m frustrated with modern tech and want to build my own satnav solution.

One limitation I have is that my car only has one USB port to benefit from the car audio system and infotainment. I’ve chosen to give the USB port to an MP3 player with my music on it.

My idea is to then get a Raspberry Pi 5 or something equivalent , probably the Pi for the community resources for the satnav system.

Add a GPS receiver to it, a generic phone screen, a few physical buttons, maybe bluetooth dongle to connect a bluetooth speaker and potentially a foldable keyboard to type addresses and install something like BRouter for local satnav. Try to figure out how to add physical buttons for media control and also manual brightness.

I’d power it with external powerbanks. The screen would be the size of a phone, or maybe even and old phone or something, to benefit from the third party market of phone holders.

The goal is relatively simple: Local offline satnav with rerouting. Full control of the data, updates and tech used. Portable so it easily comes with me from car to car over the years. Modular, so I could potentially add stuff like rear cam later on.

Why not get a dedicated GPS device? Because I don’t want to rely on a greedy corporations when I think I can do it myself (Garmin recently pulled a bad prank with a new subscription plan for instance.) And it’s simply just fun to attempt a project like this.

I have plenty of free time to learn and figure it out, but if there’s something obvious that I missed and makes the project a no-go, I’d love to know before I purchase everything.

Any feedback?

UPDATE 1st June: I’m going forward with the project. I’ve been looking extensively at how on Earth I am going to power this and the Raspberry Pi 5 isn’t a good contender because it requires 5V/5A which is very difficult to comply with in a car without tinkering that I deem advanced. I’m now considering using a Pi4. Checking if the 4 is strong enough for satnav and music.

  • Please add a disclaimer to the documents stating it was machine translated. Machine translation can get it wrong or take liberties, make up stuff. Please inform your readers so they can be on the lookout.

    Keep in mind the translated stuff by machine translation won’t be 100% what you say in your native language to other students. Be careful not to spread wrong information or knowledge.

  • Translator here. They do make up stuff or omit stuff they don’t like. Machine translation is fine for tourists or to translate a ikea manual in the wrong language. If there are stakes, risky. They got good enough to make sentences that look right so it can be tricky to spot the errors if you don’t pay attention.

    Numbers are typical errors. Sometimes it’s there but the number has changed. Sometimes it’s not there at all. Oh and if you have currencies a translators knows a document from the UK in pounds that is adapted for France will have to be converted in euros. Machines don’t.

    Generally speaking when a client wants to use machine translation, it costs them more money in the end because of the extra time needed to correct everything to a high human grade standard.