Mirror
/
ardupilot
mirror of https://github.com/ArduPilot/ardupilot
5
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
47,885 Commits 70 Branches 507 Tags 662 MiB
Commit Graph

642 Commits

Author SHA1 Message Date
Andrew Tridgell e346cd3c80 GCS_MAVLink: fixed use of incorrect millis16() values 2020-10-12 13:48:15 +11:00
Peter Barker 45f7322af6 GCS_MAVLink: avoid catching up on sending messages when sending disrupted 
The code here was never meant to maintain an "average" streamrate.  It
was designed so that we would maintain a consistent clock in the face of
minor scheduling anomalies (like an EKF fusion step).

The way this is written, however, makes us spit out a message for each
of the intervals we missed - clearly not intended behaviour.

This was tested by inserting the following code:

 void GCS_MAVLINK::update_send()
 {
+    const uint32_t xnow = AP_HAL::millis();
+    if (xnow > 10000 &&
+        xnow < 20000) {
+        return;
+    }
+
 2020-10-12 13:48:15 +11:00
Peter Barker 7cb8cde417 GCS_MAVLink: move orderly rebooting code from GCS into AP_Vehicle 
Several places we reboot the vehicle we should probably do several of
the things done in this code - flushing parameters, forcing safety on
etc.
 2020-09-29 18:07:59 +10:00
yaapu e7a49dd624 GCS_Mavlink: refactor zero_rc_outputs() out of GCS_Mavlink 2020-09-29 18:07:59 +10:00
Andy Piper 5d487be242 GCS_MAVLink: squash enum type error 2020-09-22 09:37:02 +10:00
Andy Piper 8b766efb0a GCS_MAVLink: add support for updating OSD parameters over mavlink 2020-09-09 20:36:42 +10:00
Sharvashish Das 80e2dbc9f2 GCS_Common: send_local_position uses offset from origin 2020-09-01 10:27:18 +10:00
Andrew Tridgell 30f9fd9d2a GCS_MAVLink: fixed overrun of text passed to other libraries 
this caused corruption in log files, and was passing a
non-nul-terminated string to notify and other libraries
 2020-08-28 17:35:17 +10:00
Siddharth Purohit 42f5f986d8 GCS_MAVLink: use AP_CANManager library 
also add support for handling enumeration cmd by KDETest
 2020-08-19 17:31:09 +10:00
Peter Barker 0e284b95b0 GCS_MAVLink: move battery type enumeration out of parameters object 2020-08-11 13:21:52 +10:00
Randy Mackay 1a3ef62b84 GCS_MAVLink: add virtual send_winch_status 2020-08-07 21:55:07 +09:00
elhayra 9b3a574775 GCS_MAVLink: sending UAVCAN MAVlink ESC messages 2020-08-06 09:28:35 +10:00
Andy Piper e91ae19544 GCS_MAVLink: make mount/gimbal configurable on a per-board basis 2020-08-04 09:18:14 +10:00
Dr.-Ing. Amilcar do Carmo Lucas 7067cbdafd GCS_Mavlink: Adapt to upstream mavlink changes to the SCALED_PRESSURE message 2020-07-29 14:34:22 +10:00
Michael du Breuil 6773821b3b GCS_MAVLink: Don't forward MAVLink data on channels marked private, make private channels more private 2020-07-22 07:17:49 +10:00
Andy Piper 05416daef3 GCS_Mavlink: output fast sampling startup messages 2020-07-10 16:45:29 +10:00
Andrew Tridgell 5a18b16f69 GCS_MAVLink: use strncpy_noterm 2020-07-10 13:32:24 +10:00
Peter Barker 741520d598 GCS_MAVLink: find next bucket to send when sending-ids emptied on msg removal 
We need to find the next bucket of messages to send when we've removed
the last message id from the current set of IDs we're sending out - not
just when we've empted the bucket the messages came from in the first
place.

An internal check caught this bug - the
'next_deferred_bucket_message_to_send called on empty bucket' error
 2020-07-07 10:49:31 +10:00
Peter Barker e2056f56e3 GCS_MAVLink: support for sending generator message 2020-07-03 08:21:42 +09:00
Randy Mackay 03ea646e50 GCS_MAVLink: send_battery_status uses memcpy for cell voltages 
Co-authored-by: WickedShell <Wicked.Shell.Scripts@gmail.com>
 2020-06-24 10:19:15 +09:00
Randy Mackay 65aea0e9c1 GCS_MAVLink: send_battery_status sends min of cell 10 11 and 12 2020-06-24 10:19:15 +09:00
Michael du Breuil 7fae084793 GCS_MAVLink: Only send a single battery status per call 2020-06-23 09:18:56 +10:00
Peter Barker e4621e6921 GCS_MAVLink: do not process commands after we have decided to reboot 2020-06-17 11:54:03 +10:00
chobits 991387dbfb GCS_MAVLink: support covariance field in vis pose msg 2020-06-09 13:31:21 +09:00
Peter Barker f158533fe9 GCS_MAVLink: allow private channels to also be active 
This will allow for logging of private channels.
 2020-06-09 10:37:32 +10:00
Peter Barker 06012cbc56 GCS_MAVLink: cope with RingBuffer rename if empty() to is_empty() 2020-06-09 10:21:48 +10:00
Peter Barker 07e3f1d48f GCS_MAVLink: add RADIO_STATUS.rssi as an AP_RSSI telemetry source 2020-06-09 10:20:29 +10:00
chobits ff6e4c4f9a GCS_MAVLink: support VISION_SPEED_ESTIMATE 2020-06-01 17:51:24 +09:00
Peter Barker 5100c9fb8c GCS_Common: whitelist AUTOPILOT_VERSION for in_delay_callback sending 
GCSs may request this very early on in the boot process, particularly
for SITL.

If we try to send it during a delay callback then we end up dropping it
at the moment - but we'd already sent the ack in response to the
request.
 2020-05-11 15:22:38 +10:00
Andy Piper da166b256d GCS_MAVLink: add messages to Spektrum telemetry if it is enabled 2020-05-05 09:23:15 +10:00
Andrew Tridgell 7cd901fa5b GCS_MAVLink: enable ftp without filesystem 
allows for @PARAM and @SYS
 2020-04-29 07:36:55 +10:00
Peter Barker 8b0b644c11 GCS_MAVLink: correct check for payload space for COMMAND_ACK 2020-04-27 15:30:30 +10:00
Randy Mackay 6f8176c71e GCS_MAVLink: remove ahrs3 2020-04-22 14:11:17 +09:00
Peter Barker 8b9281eaa1 GCS_MAVLink: log number of times message not sent due to no space 2020-04-17 22:21:14 +10:00
Randy Mackay c39fef6c56 GCS_MAVLink: pass vision-position-estimate reset_counter to AP_VisualOdom handler 2020-04-14 10:28:03 +10:00
Randy Mackay 9769f08fd9 GCS_MAVLink: visual odometry build fixes 2020-04-09 19:41:08 +09:00
Randy Mackay a1ddd55ea5 GCS_MAVlink: AP_integrate of rename to handle_vision_position_delta_msg 2020-04-09 19:41:08 +09:00
Randy Mackay 728e8b45a9 GCS_MAVLink: use AP_VisualOdom to handle vision-position-estimate messages 2020-04-09 19:41:08 +09:00
Siddharth Purohit b0cbbb4bea GCS_Common: fix unannotated fall-through between switch labels 2020-04-01 17:10:36 +11:00
Peter Barker 0b756547be GCS_MAVLink: add stream_slowdown to MAV message 2020-03-31 11:13:11 +11:00
Peter Barker a09ac895e9 GCS_MAVLink: remove dead assignment 
This value is never used
 2020-03-31 10:38:13 +11:00
Andrew Tridgell 8a84dbc19e GCS_MAVLink: slow down telemetry during ftp 2020-03-28 15:28:03 +11:00
Peter Barker 97b07d9c84 GCS_MAVLink: add option to get the autopilot to hard-fault itself 2020-03-19 16:03:58 +11:00
Andrew Tridgell 39fc324854 GCS_MAVLink: convert to new get_rpm() API 2020-03-12 08:36:40 +11:00
Andrew Tridgell aaf828c9d4 GCS_MAVLink: consider NO_CHANGE as success for bootloader flash 
this prevents GCS users from seeing "update failed"
 2020-03-11 16:30:06 +11:00
Rishabh f3e66055e5 GCS_MAVLink: Alert GCS if user tries calibration while vehicle is armed 2020-03-03 11:37:19 +11:00
Peter Barker fbe2e75b9f GCS_MAVLink: divide time allowed to send messages fairly 2020-02-27 08:49:59 +11:00
Peter Barker e2c9f6a8a1 GCS_MAVLink: add flags to MAV dataflash log 
holds various pieces of simple information about a link
 2020-02-25 10:59:22 +11:00
Peter Barker f8f4ed7ee0 GCS_MAVLink: log disarm method 2020-02-22 12:16:29 +11:00
Andrew Tridgell 20ea1a3dc7 GCS_MAVLink: don't send statustext from other than main thread 
This fixes high stack usage in the RCIN thread when it notifies the
GCS of a new RCIN protocol. The problem is severe when signing is
enabled, as signing adds over 500 bytes to stack cost of sending a
mavlink msg

fixes issue #13615
 2020-02-20 10:39:29 +11:00
Michael du Breuil 1f8c030eba GCS_MAVLink: Allow scripting to start/stop the REPL 2020-02-18 17:23:22 -07:00
Michael du Breuil 296c014d14 GCS_MAVLink: Fix excess trimming of autopilot version 2020-02-19 09:10:02 +11:00
Matt Lawrence c84dcd9bf8 GCS_Mavlink: Add routing for GoPro heartbeat 2020-02-18 11:13:27 +11:00
Peter Hall b1742b4e19 GCS_Mavlink: constrain battery current to avoid wrap 2020-02-12 08:04:39 +11:00
Randy Mackay 7a5f57f940 GCS_MAVLink: send rc output mode banner to GCS 2020-02-04 12:23:33 +11:00
Peter Barker 1650979a2b GCS_MAVLink: add support for MAV_CMD_DEBUG_TRAP 2020-01-29 10:27:01 +11:00
Peter Barker e248bdfb16 GCS_MAVLink: split statustexts into 50-byte chunks using mav2 extensions 2020-01-28 19:56:38 +11:00
Peter Barker 8fc64ccbe9 GCS_MAVLink: allow NavEKFs to be compiled out 2020-01-21 11:53:18 +11:00
Peter Barker a9f62f196e GCS_MAVLink: consolidate places we snprintf statustexts 2020-01-14 10:38:43 +11:00
Randy Mackay 188fdfb1b8 GCS_MAVLink: send optflow message even if no height estimate 2020-01-06 11:36:43 +09:00
Andrew Tridgell 1b528137f7 GCS_MAVLINK: cope with AP_Frsky_Telem being created after init 
may be created by the FPort backend
 2020-01-02 09:52:57 +11:00
Andrew Tridgell c6322c7c98 GCS_MAVLink: send abs throttle in VFR_HUD 2019-12-24 15:14:03 +11:00
Peter Barker 01710c035e GCS_MAVLink: handle proximity and rangefinder messages 2019-12-24 12:08:54 +11:00
Oliver Walters 5fa54f3986 GCS_MAVLink: added support for Piccolo CAN ESCs 2019-12-22 09:48:40 +11:00
Andrew Tridgell dd73eac08c GCS_MAVLink: support MAV_CMD_FIXED_MAG_CAL_YAW 2019-12-18 17:20:37 +11:00
Peter Barker ebbcd00873 GCS_MAVLink: send statustext to all GCSs until update_send is called 
until update_send it called we don't mark channels as streaming, so
send_statustext won't send to that channel.

Calculating the streaming mask could have been done as part of the init
call, but this fix may allow for information getting to the user in the
case that the streaming parameters are all zero, too.
 2019-12-18 14:36:58 +11:00
Andrew Tridgell f8f591c7db GCS_MAVLink: fixed critical error on prefight cal ops 2019-12-13 09:30:48 +11:00
Peter Barker 314ea32ede GCS_MAVLink: any internal error means MAV_STATE_CRITICAL 2019-11-27 14:27:53 +11:00
Marian Soban 6e961b55eb GCS_MAVLink: LTM telemetry support 2019-11-26 12:14:13 +11:00
Andrew Tridgell 2b27cd2ce9 Plane: Add EFI Support 2019-11-17 20:10:54 +11:00
Andrew Tridgell 00e3344321 GCS_MAVLink: add EFI_STATUS message 2019-11-17 20:10:54 +11:00
Peter Barker 7d2557b316 GCS_MAVLink: correct txspace return value issues 
- checking of space in send_to_active_channels was incorrect - did not
take into account locked status of the channel

 - corrected return value on comm_get_txspace - took a uint32_t, cast it
to int16_t, checked it for zero, then cast it to uint16_t on return.
That's just... odd.
 2019-11-12 16:33:12 +11:00
Peter Barker a893a5483c GCS_MAVLink: support sysid targetting 2019-11-12 07:52:22 +08:00
Peter Barker ac7111f1d1 GCS_MAVLink: adjust for renaming of RangeFinder files 2019-11-12 07:14:37 +08:00
Peter Barker 402f3ec09c GCS_MAVLink: honour coordinate altitude frames for DO_SET_ROI 
Also use {} to construct the location.

Note the assumption that the frame is relative-to-home when we receive
the command via COMMAND (as opposed to COMMAND_LONG)

Note that this is a behavioural change as
 2019-11-05 19:55:45 +11:00
Peter Barker b9dbfff0d0 GCS_MAVLink: announce capabilities compass-cal and param float for all vehicles 2019-11-05 11:26:23 +11:00
Michael du Breuil 6f538dd45a GCS_MAVLink: Initial FTP support 2019-11-02 22:52:06 +11:00
Peter Barker 34eff6d1c6 GCS_MAVLink: adjust for proximity status namespace change 2019-10-31 11:58:15 +08:00
Peter Barker 5c3a0926a1 GCS_MAVLink: adjust for proximity rangefinder types changing namespacing 2019-10-31 11:58:15 +08:00
Andrew Tridgell 4593697634 GCS_MAVLink: use new flash_bootloader() API 2019-10-25 21:36:55 +11:00
Michael du Breuil cb1b236439 GCS_MAVLink: Use the vehicle singleton to directly set the flight mode 2019-10-25 08:48:22 +08:00
Andrew Tridgell 29e65a7294 GCS_MAVLink: supply IMU temperature in mavlink extensions 2019-10-10 13:23:03 +11:00
Do Carmo Lucas dc4ad26e96 GCS_MAVLink: updates to support the latest stable mavlink changes (IMU temperature extension) 2019-10-10 13:23:03 +11:00
Randy Mackay 3a113a31bb GCS_MAVLink: minor fix to set_message_interval 2019-10-02 16:09:44 +09:00
Tatsuya Yamaguchi 9a99d9bff5 GCS_MAVLink: add set_message_interval functions 2019-10-02 16:09:44 +09:00
Peter Barker 6d0c5da84a GCS_MAVLink: allow upload of fence using mission item protocol 
GCS_MAVLink: add support for fence inclusion circles

GCS_MAVLink: factor out a transfer_is_complete; start commenting properly
 2019-09-28 07:58:48 +09:00
Randy Mackay 5e3c6b4d3a GCS_MAVLink: allow proximity to send only upward facing distances 
this removes the requirement that a horizontal proximity sensor be enabled
 2019-09-25 15:22:13 +09:00
Peter Barker 2f60b230cd GCS_MAVLink: use singleton to get AP_AdvancedFailsafe pointer 2019-09-17 09:13:43 +10:00
Peter Barker 8da978b913 GCS_MAVLink: use sending_mavlink1 method in send_rc_channels_raw 2019-09-12 16:10:23 +10:00
Andrew Tridgell 98c2606c0a GCS_MAVLink: fixed assumption on return of snprintf 2019-09-12 08:43:31 +10:00
Peter Barker 8fc55b9678 GCS_MAVLink: remove comm_get_available 
available returns an unsigned integer, so this was just weird
 2019-09-10 15:45:27 +10:00
Peter Barker df5921a87c GCS_MAVLink: gate AP_DEVO_Telem behind !HAL_MINIMIZE_FEATURES 2019-09-04 17:47:33 +10:00
Michael du Breuil e5eeb6181d GCS_MAVLink: Remove the usage of force_safety_no_wait 2019-09-04 07:14:50 +10:00
Michael du Breuil a74f00d045 GCS_MAVLink: Remove single use method 2019-09-03 14:35:17 +10:00
Peter Barker 5d9b332bcb GCS_MAVLink: log exectuted COMMAND_LONG and COMMAND_INT packets 2019-08-27 09:32:00 +10:00
Andrew Tridgell cfc37f2089 GCS_MAVLink: log corrected timestamp for vision data 
this allows for debugging of link lag issues
 2019-08-26 12:27:31 +10:00
Peter Barker 052e721622 GCS_MAVLink: tweak timings debug messages 2019-08-24 15:33:50 +10:00
Peter Barker 74670a77be GCS_MAVLink: avoid allocating a GCS_MAVLINK per mavlink channel 2019-08-14 18:25:43 +10:00
Peter Barker 67bf00c1c3 GCS_MAVLink: ensure payload space before sending message ack 
Also add comment about how we shouldn't be sending this message at all.
 2019-08-13 16:48:17 +10:00
Peter Barker bb542ca39a GCS_MAVLink: queue AUTOPILOT_VERSION for sending instead of direct-send 
Also allows the message to be requested for streaming or with
REQUEST_MESSAGE

We weren't checking for buffer space before sending this out, so it was
a possible source of corruption.
 2019-08-13 16:29:45 +10:00