HAL_ChibiOS: add support for PilotGaeaSH7V1-bdshot

[PilotGaea2001]

Summary

Add board support for the PilotGaeaSH7V1-bdshot flight controller.

Testing (more checks increases chance of being merged)

• Checked by a human programmer
• Tested in SITL
• Tested on hardware
• Logs attached
• Logs available on request
• Autotest included

Description

This PR adds support for the PilotGaeaSH7V1-bdshot, a high-performance STM32H743-based flight controller designed for advanced drone applications.

Key Hardware Specifications:

• MCU: STM32H743VIT6 (2MB Flash, 480MHz)
• IMU: Dual ICM42688P (on SPI1 and SPI4) for redundancy.
• Baro: DPS310 (on I2C2).
• OSD: AT7456E.
• PWM/DShot: 11 outputs total.
  • Motors 1-8: Assigned to dedicated DMA streams to ensure robust Bi-directional DShot (RPM telemetry) support.
• Other Features: Dual camera switching via GPIO, integrated 8V VTX BEC control, and microSD support.

Test Environment & Results:

• Firmware Version: Tested on latest master tag (Copter-4.6.3).
• Hardware Verification:
  • Successfully flashed and connected to Mission Planner.
  • All onboard sensors (Dual IMUs, Barometer) initialized and calibrated correctly.
  • Verified DShot functionality on Motors 1-4.
  • Confirmed peripheral functions including Camera switching and VTX power control.

*Note: Bootloader binary files are included in Tools/bootloaders/ to satisfy the AP_BOOTLOADER_FLASHING_ENABLED requirement for the initial build validation.

[PilotGaea2001]

Hi everyone, I've fixed the linting and commit naming issues. The PR is now ready. Thank your!

[Hwurzburg]

needs commits split...one per library...ie Tools: add... and AP_HAL_ChibiOS:addd.....

[PilotGaea2001]

Hi, I have split the commits as requested. One for HAL_ChibiOS and one for Tools. Ready for review, thanks!

[Hwurzburg]

Preliminary review....wont even look at README until these are addressed/discussed:

• CAN connector has USB power...almost assures bench setup failure due to excessive CAN device draws for most CAN units or failure of the CAN device...should be 5V
• Your hwdef has superfluous comments...AND multiple redefintion of the outputs...luckily our CI and Waf configure actually accepts redefinition of those for some reason.... (we need to fix this hole in CI)....but remove them anyway
• You have UART4 protocol as RCIN protocol, it should be DISPLAYPORT and you cant have two protocol 23 ports at the same time
• You set the baud rates for serial ports...no need, remove
• Since you want a serial LED setup....add the SERVO13_FUNCTION 120 in a defaults.parm file and add define DEFAULT_NTF_LED_TYPES 455 in hwdef
• The GPS pad row does not have access to I2C or 4V5 convenient. I would swap UART1 and UART2 protocols in the hwdef and update the pinout images...better wiring and convenience
• You could add booting from SD card to the bootloader def
• Since you don't pin out UART3 you could replace it in SERIAL ORDER with EMPTY to avoid giving the user a SERIAL4 parm set that does nothing for him

I will review the README once the hwdef issues are resolved or discussed....clean up the hwdef of superfluous comments/lines

[PilotGaea2001]

Thank you for the detailed feedback!
I completely understand the concerns regarding CAN power safety and the redundant definitions. I will update the hwdef.dat to swap UART1/2 for better GPS/I2C layout, clean up the superfluous comments, and fix the serial protocols/baud rates as suggested. I will also add the defaults.parm for the LED setup. I'll push the updates once I've addressed all points. Thanks!

[PilotGaea2001]

@Hwurzburg

Preliminary review....wont even look at README until these are addressed/discussed:

• CAN connector has USB power...almost assures bench setup failure due to excessive CAN device draws for most CAN units or failure of the CAN device...should be 5V
• Your hwdef has superfluous comments...AND multiple redefintion of the outputs...luckily our CI and Waf configure actually accepts redefinition of those for some reason.... (we need to fix this hole in CI)....but remove them anyway
• You have UART4 protocol as RCIN protocol, it should be DISPLAYPORT and you cant have two protocol 23 ports at the same time
• You set the baud rates for serial ports...no need, remove
• Since you want a serial LED setup....add the SERVO13_FUNCTION 120 in a defaults.parm file and add define DEFAULT_NTF_LED_TYPES 455 in hwdef
• The GPS pad row does not have access to I2C or 4V5 convenient. I would swap UART1 and UART2 protocols in the hwdef and update the pinout images...better wiring and convenience
• You could add booting from SD card to the bootloader def
• Since you don't pin out UART3 you could replace it in SERIAL ORDER with EMPTY to avoid giving the user a SERIAL4 parm set that does nothing for him

I will review the README once the hwdef issues are resolved or discussed....clean up the hwdef of superfluous comments/lines

In response to point 1 : Power distribution as per image. The CAN Bus connector mainly connects to the electronic compass. For USB-powered operation, 4.5V (VBUS via diode, actual 4.7-4.8V) is used for the CAN Bus. A redundant 5V CAN Bus path (parallel signals) is also reserved in the layout.

[PilotGaea2001]

@Hwurzburg
Thank you for the comprehensive review. I have updated the hwdef.dat and added a defaults.parm to address all the points raised:
1.The CAN Power and the 4V5 rail：

The hardware design utilizes a Dual-Diode Power Path Management (as shown in the attached schematic).Use Schottky diodes to isolate the USB VBUS and the 5V Buck Converter output.This prevents potential back-powering issues.
2.Code Cleanup: Removed all redundant definitions and superfluous comments. The hwdef.dat is now much cleaner and follows the standard ArduPilot structure.
3.Corrected UART4 protocol to DISPLAYPORT and resolved the RCIN protocol conflict.
4.Removed unnecessary default baud rate settings for all serial ports.
5.Added define DEFAULT_NTF_LED_TYPES 455 to hwdef.dat.Created etc/defaults.parm with SERVO13_FUNCTION 120 and linked it in the hwdef.
6.Swapped UART1 and UART2 protocols to ensure the GPS pad row has convenient access to I2C/4V5.
7.Enabled SD card booting in the bootloader definition.
8.Updated SERIAL_ORDER to replace the un-routed UART3 with EMPTY to simplify the parameter list for users.completely removed the unused SPI3 (pins and bus definitions) to further free up DMA and CPU resources.
The CI build is green. Please take another look at the updated implementation.

[andyp1per]

LGTM

[Hwurzburg]

0ne more nit, have approved since you respond so quickly, and added devcall label

[Georacer]

This has been included in 4.7.0-beta6. Thanks!