Path Planning Package

The path_planning package contains comprehensive set of algorithms and utilities for implementing various path planning strategies in robotic applications. This package includes refined implementations of A* (A-star), Grid Sweep, and Rapidly-exploring Random Tree (RRT) algorithms, all tightly integrated with ROS 2 for seamless deployment and visualization

Source Files and Headers

Note: File names might have changed since the initial lectures, as this represents the final form of the package as it evolved by the last lecture.

1. path_planning.cpp (Improved Version)
   • Purpose: Acts as the central node integrating all path planning algorithms (A*, RRT, Grid Sweep) and provides a unified interface for executing them based on the scenario.
   • Key Functions:
     • void PathPlanning::choose_algorithm() - Allows selection of the most appropriate path planning algorithm based on the current environment and task.
     • void PathPlanning::execute_algorithm() - Executes the selected path planning algorithm and outputs the resulting path.
     • Improvements: Enhanced logic for better decision-making when choosing the path planning algorithm, optimizing the execution flow, and integrating more robust error handling mechanisms.

Launch Files

1. path_plan_rviz.launch.py
   • Purpose: Launches the occupancy grid and path planning nodes along with RViz for visualization.
   • Nodes Launched:
     • occupancy_grid_node - Manages the occupancy grid for path planning.
     • path_planning_node - Executes the chosen path planning algorithm.
     • rviz2 - Visualizes the environment, occupancy grid, and planned path.
   • Command to run:

   ros2 launch path_planning path_plan_rviz.launch.py
   

Algorithms Implemented

1. A* (A-star)
   • Description: A grid-based algorithm that efficiently finds the shortest path using heuristic search techniques.
   • Use Case: Best suited for structured environments with well-defined grids, such as indoor navigation.
2. Grid Sweep
   • Description: Ensures complete coverage of a grid by systematically sweeping through all areas, ensuring no spot is left unvisited.
   • Use Case: Ideal for applications requiring full coverage, such as vacuum cleaning robots or agricultural drones.
3. RRT (Rapidly-exploring Random Tree)
   • Description: A versatile algorithm that quickly explores complex, high-dimensional spaces, making it suitable for environments where grid-based methods are impractical.
   • Use Case: Best for unstructured environments where traditional methods struggle, such as outdoor or dynamic environments.

CMakeLists Information

The CMakeLists.txt file is configured to build and link all components of the path_planning package:

• Executables:
  • occupancy_grid_node
  • path_planning_node
• Libraries:
  • algo_astar
  • algo_grid_sweep
  • algo_rrt
• Tests:
  • GTest-based unit and integeration tests for RRT implementations, ensuring algorithm robustness and reliability.

Building and Running

To build the package:

colcon build --packages-select path_planning

To source the setup file:

source install/setup.bash

To run the tests:

colcon test --packages-select path_planning

Example Commands

• Path Planning with RViz:

  ros2 launch path_planning path_plan_rviz.launch.py
  

Test your Learnings with Assignments

• Assignment