This project features a custom global planner for the ROS2 Navigation Stack (Nav2) based on the Dijkstra algorithm. Developed as a ROS2 plugin, it computes optimal paths in Gazebo simulations by leveraging costmap data and integrates smoothly with Nav2’s planning framework. The planner enables reliable autonomous navigation while demonstrating how classical algorithms can be applied within ROS2.
- Custom Dijkstra-based global planner plugin for Nav2
- Full integration with ROS2 Nav2 costmaps and behavior trees
- Tested in Gazebo simulation with static obstacles
- Configurable parameters for tuning planning behavior
- Modular plugin implemented using ROS2 pluginlib
- Supports end-to-end robot navigation from planning to execution