Autonomous Navigation with RRT* and A* Path Planning in ROS 2

Motion Planning, Autonomous Navigation — State Estimation, Global Planning, Control

📋 Project Overview

Architected a ROS 2 autonomous navigation pipeline with clear separation between state estimation (pose), global planning, and control, enabling modular testing and parameter tuning.

Implemented A* on an inflated occupancy grid for baseline optimal planning and RRT* for continuous-space, sampling-based motion planning in cluttered environments. Developed a pure-pursuit path tracking controller with lookahead distance, proportional gains, and velocity saturation, validating performance via planned vs. executed trajectory analysis in RViz.

⚡ Key Highlights

Modular Architecture: State estimation, global planning, and control cleanly separated
A* Planner: Optimal path planning on inflated occupancy grid
RRT* Planner: Sampling-based motion planning for cluttered environments
Pure-Pursuit Controller: Lookahead distance, proportional gains, velocity saturation
Validation: Planned vs. executed trajectory comparison in RViz

Skills Demonstrated

ROS 2 A* RRT* Motion Planning Pure-Pursuit Occupancy Grid RViz Autonomous Navigation Path Planning

📺 RViz Simulation Results

Planned vs. executed trajectory in three different environments. Green = path planned (A*/RRT*), Red = path travelled.

Environment 1

Environment 2

Environment 3