Copter: add GPS_HACC_GOOD parameter for horizontal accuracy pre-arm c…

[muramura]

Summary

Adds GPS_HACC_GOOD parameter (default: 1.5m) to ParametersG2 for Copter to enforce a maximum GPS horizontal accuracy requirement during pre-arm checks.

Classification & Testing (check all that apply and add your own)

• Checked by a human programmer
• Non-functional change
• No-binary change
• Infrastructure change (e.g. unit tests, helper scripts)
• Automated test(s) verify changes (e.g. unit test, autotest)
• Tested manually, description below (e.g. SITL)
• Tested on hardware
• Logs attached
• Logs available on request

Testing Details:

• SITL (Copter):
  • Verified default parameter is loaded as GPS_HACC_GOOD = 1.5m.
  • In LOITER mode with simulated error SIM_GPS1_ACC = 2.5m (exceeding threshold), pre-arm check failed with PreArm: High GPS HACC and prevented arming.
  • Relaxing GPS_HACC_GOOD to 3.0m cleared the error to pre-arm good and allowed arming.
  • Setting GPS_HACC_GOOD = 0 disabled the check and allowed arming.
• Hardware Test:
  • Built and verified parameter registration and MAVLink telemetry on ESP32-S3 (esp32s3m5stampfly).

Description

Motivation

• In high-precision operations (such as agricultural spraying drones using Network RTK), verifying actual positional accuracy (HAcc in meters) is critical to ensure accurate route execution and avoid boundary errors.
• While GPS_HDOP_GOOD checks satellite geometric dilution, checking horizontal_accuracy provides direct assurance that position estimate has converged before arming.
• Adding GPS_HACC_GOOD allows users to enforce a required horizontal accuracy threshold before arming (or set to 0 to disable).

Implementation Details

• Uses existing AP_GPS::horizontal_accuracy(float &hacc) method. If the connected GNSS receiver does not support/report accuracy, the check is safely bypassed.

Parameter Details

• Name: GPS_HACC_GOOD
• Class: ParametersG2
• Default: 1.5 (meters)
• Range: 0 to 10 (0 disables check)
• Increment: 0.1
• User: Advanced

AI Disclosure

This contribution was developed with AI pair-programming assistance (Antigravity).