Unable to move past 'pending' screen

[therealmoss]

I've got my Terra built and I've successfully tested all the components with the included tests in the Sensor Example Code folder.

I've not been able to get it to move past the swirl 'Pending' screen, and based on my testing with the code, there seems to be some sort of issue with the logic in the code that's preventing it from moving past this. I'm just wondering how much testing has been done with the device using real GPS and not just serial lat/long inputs?

I will continue to tinker to try and get it to work correctly and add updates to this issue.

[Moiz-I]

Did you end up fixing it and getting Terra to work correctly?
I'm deciding whether to buy the components and build it, it would help to know if the code will actually work for it haha.

[therealmoss]

Hey @Moiz-I – I have yet to get it to work unfortunately. I spent some more time getting the code to log when it reached different stages, but ran out of patience in the end. I plan to come back to it in the next couple of weeks, but ideally @modem-works need to at least let us know that they've had it working in the wild with the suggested hardware.

I'd probably hold off on the build if you're keen to get out and walk with it – I'm designing my own casing for it in the meantime!

[bryanhpchiang]

@therealmoss Were you able to make any progress here?

[therealmoss]

Hey @bryanhpchiang – not yet, but hoping to get back onto it soon. Not having much luck with @modem-works helping out here :/

[Moiz-I]

Maybe @a-l-i-c-3 can help out here?

[WannabeSpaceman]

From what I've been able to test so far, it seems like the NavigationState enum variable on line 78 of the Terra.ino file doesn't appear to get updated after its initialisation - Setting it to NAVIGATING enabled the device to skip past the pending screen, and show an arrow glyph that rotated to point towards a fixed heading, regardless of whether the debugMode bool on line 3 was set to true or false

[therealmoss]

Hey @WannabeSpaceman – yes, this is where I'm up to also. I'm confident that all my components are working correctly. Going to work through a series of tests to try and debug over the next few weeks.

[WannabeSpaceman]

@therealmoss Hey there, have you managed to make any progress on the debugging? I've had to put the Terra aside for a while to work on other projects, but now I've got the time to take another look at it, I was wondering how your own progress is coming along

[therealmoss]

@WannabeSpaceman No further than before, but I'm also going to have some time to look over the next couple of weeks. Lmk if you need me to test anything.

[xenwow]

I think @modem-works should really chime in. I've sent them an email in regards to the device and the issue threads. i also asked if they have any demo videos. We'll see what happens.

[xenwow]

@therealmoss wondering if you've got it to work or if you've made any progress?

[oowze]

Have you heard anything back from @modem-works???

[xenwow]

@oowze - they replied with a nice email!

they said it was a labour of love for them and they strongly encourage the community to further develop the project. Due to this, the amount of time they have to dedicate to further development is limited. However they are actively looking into ways to progress the project (including demos and a more detailed assembly guide).

overall they encourage us to develop the project and are looking into perhaps doing a demo or a detailed assembly guide.

[therealmoss]

Hey everyone,

I've finally had some time to get back into bug fixing. As I suspected, the code was set up to work with serial lat/longs and not real GPS data. I now have it passing GPS data properly, by bodging in some code from the GPS sensor example using ChatGPT.

This means I've finally got it reliably moving past the 'Pending' screen and onto the "Please proceed to the start of the trail." state which feels like a small but important first step.

I've got it pumping data into the Serial Monitor on the IDE so that I can see what numbers it's sending out and where. Main thing I've sorted is that the 'distance' variable needed to be divided by 10000 to give an integer that represents meters. Still some tinkering to do, but I will share a branch soon for folks to take a look at and try out.

@xenwow I'm glad to hear Modem are keen, but also they could still have been much more helpful here by at least giving us some insight into what state the code was in 😅

Wish me luck as I continue to work on this! I'll share a video if I get the flipping thing actually working.

[xenwow]

@therealmoss thanks for all your effort - really!

from what I understood the way they had the device working was through that HTML site where you place your destination + duration and generate a trail.

You've now had it use live location to automatically generate a route with some kind of prompt somewhere? (sorry im not
the most techy)