Pancake: An Automated Guided Vehicle with ROS and Visual SLAM
/ 1 min read
Updated:Table of Contents
Pancake
Pancake is an open-source automated guided vehicle built as a practical robotics platform. The name comes from the goal of making Pan a piece of cake.
The project combines a Jetson Nano, a Kinect V2 RGB-D camera, ROS, and visual SLAM. Its software and hardware notes are available in the Pancake GitHub repository.
System overview
- Compute: NVIDIA Jetson Nano
- Perception: Kinect V2 RGB-D camera
- Middleware: ROS 1
- Localization and mapping: visual SLAM
- Purpose: provide a compact platform for experimenting with mapping, navigation, and robot control
The RGB-D camera supplies color and depth data to the perception pipeline. ROS nodes expose the sensor streams and connect them to the SLAM and vehicle-control components.
Related setup notes
Project status
This page is an overview rather than a complete build guide. For the tested source code, hardware details, and current limitations, refer to the repository and its revision history.