BeaverworksGit

beaverworksGit

A maze-navigating robot built around an Arduino Mega 2560, combining custom PCB design (KiCad) with firmware for grid-based autonomous navigation, sensor logging, and ESP32-CAM video streaming.

Overview

The robot drives through a 5x5 grid maze along a pre-defined path, correcting its heading with an IMU (gyro yaw tracking) instead of relying on open-loop timing alone. At each cell it pauses to read temperature, humidity, and distance-to-obstacle, then reports the data back over serial. A light sensor and buzzer provide simple flashlight/alarm behavior along the way, and a separate ESP32-CAM module streams live video.

Repository Structure

.
├── beaverworksGit.kicad_pro/.kicad_sch/.kicad_pcb  # KiCad project (schematic + PCB layout)
├── fp-lib-table                                    # KiCad footprint library table
├── hardware/                                        # Hardware design files
│   ├── beaverworksGit.kicad_*                       #   (duplicate/working copy of the KiCad project)
│   ├── Arduino Mega 2560 Rev3 A000067/               #   Arduino Mega 3D model + footprint
│   ├── Arduino_MountingHole.pretty/                  #   Mounting hole footprints
│   ├── HC-SR04 Model by SparkFun/                    #   Ultrasonic sensor 3D model + footprint
│   └── esp32-wrover-kicad-master/                    #   ESP32-WROVER KiCad library (third-party)
├── software/                                        # Firmware / companion scripts
│   ├── IMU Navigation/
│   │   ├── ObstacleAvoidance.ino                     #   Standalone obstacle-avoidance sketch
│   │   ├── Calibration/calibration.ino                #   IMU/gyro calibration routine
│   │   └── Yaw Tracking/mpu6050_copy_*/               #   MPU6050 yaw-tracking test sketch
│   ├── ESP32_Camera/                                 #   ESP32-CAM web server + streaming
│   │   ├── CameraWebServer.ino                        #     Main camera sketch (based on Espressif's CameraWebServer example)
│   │   ├── app_httpd.cpp / camera_pins.h / ...         #     Supporting camera server code
│   │   ├── wifi_secrets.example.h                     #     Template for Wi-Fi credentials (copy to wifi_secrets.h)
│   │   ├── record_video.py / stream_test.py            #     Python helpers for grabbing/testing the video stream
│   │   └── python/                                     #     Additional Python utilities
│   └── Sensors_Code/                                 #   (sensor test code)
├── main.cpp                                          # Main robot firmware: timed turns/drives, no gyro
├── navigatio.cpp                                     # Main robot firmware: gyro-corrected turns/drives (MPU6050)
└── platformio.ini                                    # PlatformIO project config (Arduino Mega target)

Hardware

• MCU: Arduino Mega 2560 Rev3
• IMU: Adafruit MPU6050 (gyro-based yaw tracking)
• Distance sensing: HC-SR04 ultrasonic sensor
• Environmental sensing: DHT11 temperature/humidity sensor
• Other I/O: LDR (light sensor), sound sensor, buzzer, status LED
• Drive: Dual DC motors via an H-bridge driver (enable + direction pins per side)
• Camera: Separate ESP32-CAM module for live video streaming
• PCB: Custom board designed in KiCad (see beaverworksGit.kicad_pro / hardware/)

Firmware

There are two versions of the main navigation firmware:

Both firmware versions:

1. Follow a hard-coded path[] array of NORTH/EAST/SOUTH/WEST moves through a 5x5 grid.
2. Stop at each cell to read temperature, humidity, and ultrasonic distance, and report them over serial.
3. Continuously monitor light and sound sensors to trigger an LED "flashlight" and buzzer alert.
4. Stop and idle once the path is complete.

Building with PlatformIO

The project is set up for PlatformIO targeting the Arduino Mega:

# Build
pio run

# Upload to the board
pio run -t upload

# Open the serial monitor (9600 baud)
pio device monitor

Dependencies (declared in platformio.ini):

• adafruit/Adafruit MPU6050
• adafruit/Adafruit Unified Sensor
• adafruit/DHT sensor library

Note: only one of main.cpp / navigatio.cpp should be treated as the active src/main.cpp for a given build — move/copy whichever version you want to flash into PlatformIO's src/ directory (or update platformio.ini's build_src_filter accordingly).

ESP32-CAM streaming

software/ESP32_Camera/CameraWebServer.ino is based on the standard Espressif CameraWebServer example, adapted with local board/pin config. To use it:

1. Copy wifi_secrets.example.h to wifi_secrets.h and fill in your Wi-Fi SSID and password.
2. Flash CameraWebServer.ino to an ESP32-CAM board via the Arduino IDE (select the correct board and partition scheme — see partitions.csv).
3. Use stream_test.py or record_video.py to test/capture the resulting MJPEG stream from a computer on the same network.

IMU navigation experiments

software/IMU Navigation/ contains standalone sketches used during development:

• Calibration/calibration.ino — computes gyro bias for the MPU6050.
• Yaw Tracking/mpu6050_copy_*/ — isolated yaw-tracking test.
• ObstacleAvoidance.ino — simple reactive obstacle-avoidance behavior.

Hardware Design (KiCad)

Open beaverworksGit.kicad_pro in KiCad to view or edit the schematic (.kicad_sch) and PCB layout (.kicad_pcb). Third-party footprints/3D models used in the design are vendored under hardware/ (Arduino Mega, HC-SR04, ESP32-WROVER, mounting holes).

Status

This is a work-in-progress class/competition project (Beaverworks). Firmware constants such as MS_PER_INCH, MS_PER_90_TURN, drive speeds, and the maze path[] are tuned for a specific physical maze layout and will need recalibrating for a different course or robot chassis.