ROS2 Omnidirectional Robot

An omnidirectional mobile robot running ROS2, with LiDAR, a depth camera and room for expansion. It does SLAM mapping, autonomous navigation, obstacle avoidance and visual tracking. It is built for robotics teaching, research and real-world uses such as inspection and mapping.

Front view of a mecanum-wheel robot on a tiled floor; a touchscreen on the front shows a mapping program with a grid, and a black depth camera sits on top — Mapping software on the onboard screen

Angled view of the same mecanum-wheel robot on a tiled floor, with a tilted touchscreen and a black depth camera on an aluminum frame — Mecanum chassis, depth camera and touchscreen

Tech stack

Main computer: Raspberry Pi 4B, Ubuntu + ROS2
Sensors: EAI G4 LiDAR, Orbbec Astra Pro depth camera, IMU, wheel encoders
Drive: 4 independent mecanum wheels, motor drivers (PWM / serial)
Mapping & navigation: SLAM (Gmapping / Karto / Hector / Cartographer), TEB / DWA navigation, A* / RRT path planning
Perception: KCF visual tracking, line following
Simulation: Gazebo
Power & frame: 12 V battery, aluminum chassis
Control: Wi-Fi mobile app

What we built

Robot chassis design
Carrier board and hardware development: sensor integration, power management, motor control
Embedded software on ROS2
System integration, debugging and testing
Delivered complete hardware, software source code, documentation, assembly drawings and a test guide

Development time

18 days

All builds →

Building something similar?

Review my project

Or email [email protected] · WhatsApp +1 (323) 719-3730