GCS_MAVLink: Add FLIGHT_INFORMATION message #25489

nexton-winjeel · 2023-11-08T06:27:18Z

Adds ~~a new FLIGHT_DURATION~~ the FLIGHT_INFORMATION message to report the time that the vehicle armed and took off (if known).

This message is needed if connecting a GCS to a vehicle that is already armed/flying, as otherwise these times can't be accurately determined.

Also adds an accessor to AP_Arming to record the arming time.

~~Requires: ArduPilot/mavlink#336~~

IamPete1 · 2023-11-08T11:11:42Z

I wonder if the new metrics would be useful as part of AP_Stats, you can get flight time from there, but currently you have to pull the param value.

peterbarker · 2024-04-05T01:14:08Z

@nexton-winjeel I think this PR is now no-longer-required as we're going to use FLIGHT_INFORMATION instead, correct?

nexton-winjeel · 2024-04-05T02:21:20Z

@peterbarker: It's not required as-is, but I'll convert it to use FLIGHT_INFORMATION instead of the now non-existent FLIGHT_DURATION. Will get to it early next week.

peterbarker · 2024-05-22T13:28:38Z

@peterbarker: It's not required as-is, but I'll convert it to use FLIGHT_INFORMATION instead of the now non-existent FLIGHT_DURATION. Will get to it early next week.

Long week?

nexton-winjeel · 2024-05-23T00:00:00Z

Long week?

Very. Still not my oldest open PR though...

peterbarker

Would be nice to get the message updates down from mavlink/mavlink.

I don't think we need to tie this with AP_Stats now. It's not a huge amount of shared state.

libraries/GCS_MAVLink/GCS_Common.cpp

peterbarker · 2024-07-29T08:29:21Z

libraries/GCS_MAVLink/GCS_Common.cpp

+    // We send time since boot (in micros) rather than a UTC timestamp, as this
+    // works better with SITL when SIM_SPEEDUP > 1


This comment is out of date once the patches to the mavlink message upstream are brought back.

Instead it should read something like the fields are misnamed...

Also, the flag value is a little problematic with the problematic ARMING_REQUIRE flag on Rover and Plane. There's always an argument that they're not really armed in that state, but just pretend to be.... but still, how does this message work for an ARMING_REQUIRE 0 vehicle?

[...] There's always an argument that they're not really armed in that state, but just pretend to be...

They will set the MAV_MODE_FLAG_SAFETY_ARMED flag of the base_mode field of the HEARTBEAT message, so as far as a GCS knows, they are armed.

[...] how does this message work for an ARMING_REQUIRE 0 vehicle?

It will report arming time as 0 if AP_Arming::arm() is never called. This field shouldn't be treated as a proxy for "is armed", it needs to be read in conjunction with the arming state reported in the HEARTBEAT.

libraries/AP_Arming/AP_Arming.h

libraries/GCS_MAVLink/GCS.h

peterbarker · 2024-07-31T03:24:30Z

Closes #25284

peterbarker · 2024-08-03T01:27:09Z

libraries/GCS_MAVLink/GCS_Common.cpp

    float current, consumed_mah, consumed_wh;
    const int8_t percentage = battery_remaining_pct(instance);
-    
+


Please remove the whitespace changes.

Fix the whitespace problems as you change the code.

peterbarker

Need to fix that type, but LGTM otherwise.

peterbarker · 2024-08-06T05:09:18Z

libraries/GCS_MAVLink/GCS_Common.cpp

+#if AP_MAVLINK_MSG_FLIGHT_INFORMATION_ENABLED
+void GCS_MAVLINK::send_flight_information()
+{
+    const uint32_t time_boot_us = AP_HAL::millis64();


Suggested change

const uint32_t time_boot_us = AP_HAL::millis64();

const uint64_t time_boot_us = AP_HAL::millis64();

peterbarker · 2024-08-06T05:10:22Z

libraries/GCS_MAVLink/GCS_Common.cpp

+        switch (current_landed_state) {
+            case MAV_LANDED_STATE_IN_AIR:
+            case MAV_LANDED_STATE_TAKEOFF:
+            case MAV_LANDED_STATE_LANDING: {


Don't really need to open braces if you're not declaring variables.

Doesn't need to be fixed, just maybe leave them out in future

Defensive programming habit. Fixed while I was in there...

peterbarker · 2024-08-06T05:13:21Z

libraries/GCS_MAVLink/GCS_config.h

 #endif
+
+#ifndef AP_MAVLINK_MSG_FLIGHT_INFORMATION_ENABLED
+#define AP_MAVLINK_MSG_FLIGHT_INFORMATION_ENABLED 1


What's the flash size cost of this new feature?

Will it cause us to start overflowing on smaller boards?

Should we add it into minimize_common.inc? (@andyp1per and @Hwurzburg might have ideas on that)

We don't need to change this as part of this PR.

What's the flash size cost of this new feature?

~728 bytes (tested on CubeOrange).

Target Text (B) Data (B) BSS (B) Total Flash Used (B) Free Flash (B) External Flash Used (B) ---------------------------------------------------------------------------------------------------------- Disabled: bin/arducopter 1840268 3920 258372 1844188 121876 Not Applicable Enabled: bin/arducopter 1840996 3920 258372 1844916 121148 Not Applicable

Hwurzburg

I think this should be disabled by def and a build option...not going to be used by 95% of users....or at least disabled in minimize include...also needs file for build server and features list added

peterbarker · 2024-08-06T23:45:50Z

I think this should be disabled by def and a build option...not going to be used by 95% of users....or at least disabled in minimize include...also needs file for build server and features list added

This is going to be a good/useful message for a lot of people. I plan on patching MAVProxy for it to fix the resetting-flight-time problem it has.

@nexton-winjeel please add an entry into minimize_common.inc so this isn't built on the tiny boards. Also entries to extract_features.py and build_options.py so users can turn the feature on/off as desired.

nexton-winjeel · 2024-08-07T00:26:40Z

@nexton-winjeel please add an entry into minimize_common.inc so this isn't built on the tiny boards. Also entries to extract_features.py and build_options.py so users can turn the feature on/off as desired.

Done.

peterbarker

We need a mapping in GCS_MAVLINK::mavlink_id_to_ap_message_id for this message

Also a fix for ms vs us

Consider grabbing my patch from my pr/Winjeel/upstream/add-FLIGHT_DURATION-message branch

peterbarker · 2024-08-14T06:32:21Z

libraries/GCS_MAVLink/GCS_Common.cpp

+
+    mavlink_msg_flight_information_send(
+        chan,
+        time_boot_us,


This is actually a time_boot_ms field...

peterbarker · 2024-08-14T06:50:38Z

In-progress MAVProxy support is here: ArduPilot/MAVProxy#1438

Works ~~, but I've broken other stuff in a refactor~~

peterbarker · 2024-10-18T02:32:09Z

Ping @nexton-winjeel - might still be viable for 4.6 - but even than had better be soon :-)

nexton-winjeel · 2024-10-28T03:06:29Z

@peterbarker: Can I get you to review this again?

patrickelectric · 2025-03-13T20:31:22Z

Can we have this merged ?

peterbarker · 2025-03-16T23:27:44Z

libraries/GCS_MAVLink/GCS_Common.cpp

+    const uint64_t time_boot_micros = AP_HAL::micros64();
+    const uint32_t time_boot_ms = static_cast<uint32_t>(time_boot_micros / 1000);
+
+    // This field is misnamed as `arming_time_utc` in MAVLink. However, it is
+    // not a UTC time, it is the microseconds since boot.
+    const uint64_t arm_time_us = AP::arming().arm_time_us();


Prefer to declare/define variables as late as possible.

In this case, they're declared in the order they are used in the mavlink_msg_flight_information_send function.

peterbarker · 2025-03-16T23:32:19Z

I can't force-push this branch.

I've rebased this branch over here: #29541 - just resolved the conflicts.

I intend to merge that other PR once it passes CI. And then the MAVProxy one.

peterbarker · 2025-03-16T23:34:09Z

@peterbarker: Can I get you to review this again?

Apologies - missed this request.

…s.py

nexton-winjeel · 2025-03-17T05:43:47Z

I can't force-push this branch.

I've rebased this branch over here: #29541 - just resolved the conflicts.

I intend to merge that other PR once it passes CI. And then the MAVProxy one.

I've rebased this one if you want to merge this instead of the other branch.

peterbarker · 2025-03-17T10:34:40Z

Merged, thanks!

nexton-winjeel marked this pull request as draft April 5, 2024 02:17

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from 0becf38 to c987d57 Compare May 22, 2024 23:56

nexton-winjeel changed the title ~~GCS_MAVLink: Add FLIGHT_DURATION message~~ GCS_MAVLink: Add FLIGHT_INFORMATION message May 22, 2024

nexton-winjeel marked this pull request as ready for review May 23, 2024 00:00

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from c987d57 to faff94b Compare July 25, 2024 01:34

joshanne added the DevCallTopic label Jul 25, 2024

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from faff94b to d0af9b8 Compare July 25, 2024 01:42

peterbarker requested changes Jul 29, 2024

View reviewed changes

CraigElder removed the DevCallTopic label Jul 29, 2024

peterbarker reviewed Jul 29, 2024

View reviewed changes

libraries/AP_Arming/AP_Arming.h Outdated Show resolved Hide resolved

tridge reviewed Jul 29, 2024

View reviewed changes

libraries/AP_Arming/AP_Arming.h Outdated Show resolved Hide resolved

tridge reviewed Jul 29, 2024

View reviewed changes

libraries/GCS_MAVLink/GCS.h Outdated Show resolved Hide resolved

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from d0af9b8 to f4312c8 Compare July 31, 2024 05:57

nexton-winjeel requested review from peterbarker and tridge July 31, 2024 05:58

peterbarker requested changes Aug 3, 2024

View reviewed changes

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from f4312c8 to 3447e4f Compare August 4, 2024 23:59

peterbarker requested changes Aug 6, 2024

View reviewed changes

Hwurzburg requested changes Aug 6, 2024

View reviewed changes

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from 3447e4f to ea6e560 Compare August 7, 2024 00:25

nexton-winjeel requested a review from peterbarker August 7, 2024 00:32

peterbarker requested changes Aug 14, 2024

View reviewed changes

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from ea6e560 to 8ba89e2 Compare October 18, 2024 04:34

nexton-winjeel requested a review from peterbarker October 18, 2024 04:34

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from 8ba89e2 to bd9b7cd Compare October 18, 2024 04:38

IamPete1 added the DevCallTopic label Mar 13, 2025

peterbarker added this to Peter's ArduPilot 4.7 Queue Mar 16, 2025

peterbarker reviewed Mar 16, 2025

View reviewed changes

peterbarker mentioned this pull request Mar 16, 2025

GCS_MAVLink: Add FLIGHT_INFORMATION message #29541

Closed

nexton-winjeel added 4 commits March 17, 2025 16:15

AP_Arming: Record arming time and add accessor for it

fe6334f

GCS_MAVLink: Add FLIGHT_INFORMATION message

7996df7

AP_HAL_ChibiOS: Disable FLIGHT_INFORMATION msg in minimize_common.inc

80cfa8a

Tools: Add FLIGHT_INFORMATION to build_options.py and extract_feature…

b365113

…s.py

nexton-winjeel force-pushed the upstream/add-FLIGHT_DURATION-message branch from bd9b7cd to b365113 Compare March 17, 2025 05:41

peterbarker merged commit f328a65 into ArduPilot:master Mar 17, 2025
106 of 107 checks passed

github-project-automation bot moved this to Done in Peter's ArduPilot 4.7 Queue Mar 17, 2025

IamPete1 removed the DevCallTopic label Mar 17, 2025

peterbarker removed this from Peter's ArduPilot 4.7 Queue Mar 18, 2025

		// We send time since boot (in micros) rather than a UTC timestamp, as this
		// works better with SITL when SIM_SPEEDUP > 1

		float current, consumed_mah, consumed_wh;
		const int8_t percentage = battery_remaining_pct(instance);

	const uint32_t time_boot_us = AP_HAL::millis64();
	const uint64_t time_boot_us = AP_HAL::millis64();

GCS_MAVLink: Add FLIGHT_INFORMATION message #25489

GCS_MAVLink: Add FLIGHT_INFORMATION message #25489

Uh oh!

Conversation

nexton-winjeel commented Nov 8, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

IamPete1 commented Nov 8, 2023

Uh oh!

peterbarker commented Apr 5, 2024

Uh oh!

nexton-winjeel commented Apr 5, 2024

Uh oh!

peterbarker commented May 22, 2024

Uh oh!

nexton-winjeel commented May 23, 2024

Uh oh!

peterbarker left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

nexton-winjeel Jul 31, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

peterbarker commented Jul 31, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

peterbarker left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

nexton-winjeel Aug 7, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Hwurzburg left a comment

Choose a reason for hiding this comment

Uh oh!

peterbarker commented Aug 6, 2024

Uh oh!

nexton-winjeel commented Aug 7, 2024

Uh oh!

peterbarker left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

peterbarker commented Aug 14, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

peterbarker commented Oct 18, 2024

Uh oh!

nexton-winjeel commented Oct 28, 2024

Uh oh!

patrickelectric commented Mar 13, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

nexton-winjeel commented Nov 8, 2023 •

edited

Loading

nexton-winjeel Jul 31, 2024 •

edited

Loading

nexton-winjeel Aug 7, 2024 •

edited

Loading

peterbarker commented Aug 14, 2024 •

edited

Loading