Waypoint Estimation

Code for an autonomous scale tractor using an STM32H755 dual-core module to follow GPS waypoints and communicate with a John Deere camera. Includes C implementations for sensor integration and real-time path control using SPI, I2C, UART, and CAN protocols.

Developed by: Humberto, Roberto, Mauricio, and Erick

Overview

This project implements an autonomous waypoint estimation system using a scale tractor. The system integrates an STM32H755 dual-core microcontroller, a John Deere camera for GPS waypoint detection, and a variety of sensors and actuators to navigate a predefined route. It utilizes advanced communication protocols (SPI, I2C, UART, CAN, and wireless communication) and FreeRTOS for real-time task management.

The tractor is capable of:

• Receiving GPS coordinates from a John Deere camera.
• Using sensor data (IMU, encoder) to estimate its position and navigate accurately.
• Managing motor control (traction and steering) using PWM signals.
• Efficiently communicating data between subsystems using a modular design.

---

Files

The repository is organized into the following directories:

• Arduino: Contains all the code for the Arduino Nano, including:

  • CAN communication to send encoder data to the STM32H755.
  • Encoder interface code to measure the DC motor's rotation.

• STM: Includes all the firmware for the STM32H755, organized by core functionality:

  • M4 Core: Dedicated to auxiliary tasks, such as toggling LEDs when the tractor crosses a waypoint.
    • myprintf.c: Debugging helper for message formatting.
  • M7 Core: Handles the main operations of the tractor, including:
    • main.c: Entry point for the STM32H755 firmware.
    • freertos.c: Implementation of FreeRTOS tasks for task management.
    • spi.c: SPI communication with the nRF24 wireless module and the Deere GPS camera.
    • i2c.c: I2C communication for reading data from the IMU.
    • can.c: CAN communication to receive encoder data from the Arduino Nano.
    • steering.c: Code for controlling the servo motor for navigation.
    • mpu6050.c: Driver code for the IMU module.
    • WayPointNotify.c: State machine implementation for waypoint notification and updates.
  • Common Files: Shared between the two cores.
    • stm32h7xx_it.c: Interrupt handlers for the SysTick and other peripherals.
    • syscalls.c and sysmem.c: System-level utilities.

• LPS: Contains the Python script used to operate the John Deere LPS camera, which detects the tractor's position and sends waypoint coordinates wirelessly.

• Gallery: Includes images and videos documenting the project development process, such as:

  • System schematics.
  • Pin configurations.
  • Testing area and final product images.

---

Prerequisites

To replicate this project, you will need:

• STM32H755 dual core module
• Arduino Nano
• ESC Module
• nRF24 Wireless module
• TJA1051 CAN transceiver
• MCP2515 CAN Shield
• DC motor with encoder (Powered by a 9 V battery)
• Servo motor
• LM2596 Voltage regulator
• LPS John Deere Camera
• 3.7 V LiPo Battery
• MPU6050 IMU Module
• 9 V Battery
• Breadboards
• Jumper wires
• 4 AA Batteries (6 V)

---

Schematic

The complete schematic of the system, including all components and connections, can be found in the Gallery folder. Below is a preview:

---

Final product

The final integrated system demonstrates a fully functional scale tractor navigating a U-shaped route with waypoint notifications and updates. Below is an image of the final product:

---

Testing area

The testing area consisted of a controlled black environment monitored by the John Deere camera, which provided the GPS coordinates for the tractor's navigation. A snapshot of the testing area is shown below:

---

IOC, Pins and peripherals

The project utilized a wide array of peripherals configured in the STM32CubeMX .ioc file. These configurations include:

• SPI: For nRF24 and Deere GPS camera communication.
• CAN: For communication between STM32 and Arduino.
• I2C: For IMU data acquisition.
• UART: For debugging and serial communication.
• PWM: For motor control (traction and steering).

---

Clock configurations

The STM32H755's clock was configured for optimal performance using:

• System Clock (HCLK): Configured to 400 MHz for efficient dual-core operation.
• Peripheral Clocks: Adjusted to meet the timing requirements of SPI, I2C, and CAN modules.

Details of the clock configuration are shown below:

---

Diagrams

The project includes several development diagrams to illustrate system design:

---