ADSB: create ADS-B backend split and add Sagetech driver #15576
Conversation
There was a problem hiding this comment.
More patches required. Too lumpy.
One patch for each logical change.
A patch to create the frontend/backend split, then another that only moves the code between files.
I stopped checking through the mavlink backend part-way through as any issues in there are probably pre-existing. However, since the patch history in this PR doesn't make it clear that no code has changed I'd want to go through it with a fine-toothed comb next review.
github interface was stuffing up association of comments with lines - so things are a little wonky.
| @@ -128,6 +128,23 @@ float throttle_curve(float thr_mid, float alpha, float thr_in) | |||
| return thr_out; | |||
| } | |||
|
|
|||
| /* | |||
| * Convert any base number to any base number. Example octal(8) to decimal(10) | |||
There was a problem hiding this comment.
So... that return type looks like it might not handle hexadecimal too well.
This function would need a big warning that it's expensive.
| } | ||
| } | ||
|
|
||
| uint16_t fetchU16(const uint8_t *v, bool MSBfirst) |
There was a problem hiding this comment.
I don't think these add anything to our existing (e.g) le16toh_ptr functions.
They're certainly not of the same style: https://github.com/ardupilot/ardupilot/blob/pr%2Fbutton-pwm-and-actions/libraries/AP_Common/AP_Common.h#L88 and https://github.com/ardupilot/ardupilot/blob/pr%2Fbutton-pwm-and-actions/libraries/AP_HAL/utility/sparse-endian.h#L88
There was a problem hiding this comment.
will remove these and create UINT24_VALUE()
|
|
||
| // longitude and latitude | ||
| // NOTE: these MUST be done in double or else we get roundoff in the maths | ||
| const double lon_deg = longitude * (double)1.0e-7 * (longitude < 0 ? -1 : 1); |
| const Vector2f speed = AP::ahrs().groundspeed_vector(); | ||
| float speed_knots = norm(speed.x, speed.y) * M_PER_SEC_TO_KNOTS; |
There was a problem hiding this comment.
| const Vector2f speed = AP::ahrs().groundspeed_vector(); | |
| float speed_knots = norm(speed.x, speed.y) * M_PER_SEC_TO_KNOTS; | |
| const Vector2f speed = AP::ahrs().groundspeed_vector(); | |
| float speed_knots = speed.length() * M_PER_SEC_TO_KNOTS; |
| // time | ||
| uint64_t time_usec; | ||
| if (!AP::rtc().get_utc_usec(time_usec)) { | ||
| memset(&pkt.payload[36],' ', 10); |
| } | ||
| } | ||
|
|
||
| const char* AP_ADSB_Sagetech::systemStatsBits_to_str(const SystemStateBits systemStateBits) |
There was a problem hiding this comment.
This really doesn't do what its name hints it does - this is a bitmask and this won't return you anything useful if two bits are set.
There was a problem hiding this comment.
It should also be static
| send_msg(pkt); | ||
| } | ||
|
|
||
| const char* AP_ADSB_Sagetech::type_to_str(const uint8_t type) |
|
I'll close this PR in favor of #15610 which is a pre-backend split PR. This PR has lots of good comments from @peterbarker which I incorporated into it |
This adds a driver for the Sagetech ADSB transceiver. The ADSB wiki page is getting a little out of date so I plan to update that as well.
This backend split is needed because there is already a Sagetech MX protocol and uAvionix GDL90/UCP protocol in the pipeline.
This is a re-do of PR #15123