Differential Drive Robot system supports mapping, localization and navigation using Cartographer, Nav2, LiDAR, Raspberry Pi 4 and Arduino/ESP32 encoder communication to calculate odometry.
The robot consists of 3 main parts:
- Low-level control layer Arduino/ESP32 reads encoder, controls motors and sends data via serial.
- Compute Layer Raspberry Pi 4 runs ROS2, receives encoder data from Arduino to calculate odometry, reads LiDAR data to do SLAM, and runs Nav2 to move automatically.
- Application & Monitoring Rviz2 to observe 3D robot, trajectory, map, TF, Nav2 status,…
-
Raspberry Pi 4 (4GB or 8GB)
-
Arduino Nano / ESP32 for motor control & encoder reading
-
Motor driver L298N / BTS7960 / Cytron
-
Motor GA25 – 120RPM, 7.3kg.cm (stall), 1.8A with load
-
LiDAR (YDLIDAR X3 Pro or RPLIDAR A1/A2)
-
Battery power 12V 10C
-
Robot frame + wheels
- ROS2 uses node diff_driver.py to communicate with Arduino via serial /dev/arduino (57600 baud).
- Sen command m 100 100" to set frame rate
- Send command "o 100 100" set PWM
- Send command "e" to get encoder data. Arduino will send to Pi in form "e_left e_right"
- Send "r" command to reset encoder value
diff_driver.launch.py
- Open serial
- Send motor command (PWM + direction)
- Receive encoder
- Publish to topic: /left_speed /right_speed
Cartographer will subscribe:
- /scan from LiDAR
- /odom (Self published odom)
ros2 launch robot_mapping cartographer.launch.py
Save map
ros2 run nav2_map_server map_saver_cli -f my_map
- map server
- amcl (localization)
- controller server
- planner server
- smoother server
- BT navigator
- Lifecycle manager
ros2 launch robot_navigation naviagation.launch.py
- Name: BUI QUOC DOANH
- Email: doanh762003@gmail.com
- Project: diff robot
This project is released under the MIT License