PRM-Nav

Overview

PRM-Nav implements a Probabilistic Roadmap (PRM) navigation framework for a differential-drive robot in a planar environment with circular obstacles. The system receives goals via /goal_pose and executes smooth, collision-free trajectories while visualizing paths and obstacles in RViz.

Features

• PRM-based path planning with random sampling and k-nearest neighbor connections
• Dijkstra’s algorithm for shortest path computation
• Path simplification and smoothing to reduce unnecessary waypoints
• Obstacle handling with inflation for safe navigation
• ROS2 node (prm_navigator) integrating planning, control, and visualization
• Simulation with sim1 node for robot motion and RViz visualization
• Supports multiple environments with configurable obstacles

Usage

1. Launch the Simulator

ros2 run sim sim1

• Starts the simulation node
• Robot pose published on /pose
• Past robot positions visualized via /sim_markers

2. Launch RViz2

rviz2

• Add the following displays:

  • RobotModel: Subscribe to /robot_description
  • MarkerArray: Subscribe to /sim_markers (robot trajectory)
  • MarkerArray: Subscribe to /obstacle_markers (obstacles)
  • Path: Subscribe to /planned_path (planned PRM path)

• Set the fixed frame to world.

3. Launch the PRM Navigator Node

ros2 launch projectz navigator_launch.py obstacles:="0.0,2.0,0.8; -2.0,0.0,0.5; 2.0,-1.5,0.65; -1.5,-2.5,0.7"

• Configure different environments by changing the obstacle string
• Node subscribes to:
  • /pose → Current robot pose
  • /goal_pose → Target goal from RViz
• Node publishes:
  • /cmd_vel → Velocity commands for the robot
  • /planned_path → Planned path for visualization
  • /obstacle_markers → Obstacle visualization in RViz

4. Set Goals in RViz

• Use the “2D Goal Pose” tool
• Click a location in the world to set the goal
• Robot automatically plans and moves toward the goal

Demonstration

PRM_Nav.mp4

File Structure

PRM-Nav/
├── projectz/
│   ├── navigator.py        # Main PRM navigator ROS2 node
│   ├── prm.py              # PRM planner implementation
│   ├── geometry.py         # Geometric utilities and collision checks
├── launch/
│   └── navigator_launch.py # Launch file with configurable obstacles
├── sim1.py                 # Robot simulation node
└── README.md               # Project documentation

Results

• Successfully navigates environments with 4, 5, and 6 obstacles
• Handles straight-line, multi-turn, and unreachable goals
• Maintains safe clearance and smooth trajectories
• Path simplification and smoothing keeps paths near-optimal while avoiding collisions