Repository files navigation

This repository contains implementations of variousadvanced path-planning algorithms used in autonomous mobile robots. These algorithms are widely applied inrobotics, AI, and autonomous navigation systems.

The repository includes implementations of:

• A*,D*,Hybrid DQN-A*,RRT*, andTheta* algorithms.
• Efficient solutions fornavigation in dynamic and static environments.

✔A* Algorithm - Implements the classical A* search and bidirectional variant for optimal path planning.
✔D* Algorithm - A dynamic path planning algorithm that adapts to changing environments.
✔Hybrid DQN-A* - A hybrid approach combining Deep Q-Network (DQN) with A* for intelligent decision-making.
✔RRT* Algorithm - Rapidly-exploring Random Tree (RRT*) for sampling-based path planning.
✔Theta* Algorithm - An optimized variant of A* allowing any-angle path planning.

📦 Advanced_Path_Planning├── 📂 A_Star│   ├── 📄 README.md│   ├── 📄 a_star_algorithm.py│   ├── 📄 a_star_bidirectional_algorithm.py│├── 📂 D_Star│   ├── 📄 README.md│   ├── 📄 d_star.py│├── 📂 Hybrid_DQN_A_Star│   ├── 📄 README.md│   ├── 📄 hybrid_dqn_a_star_algorithm.py│├── 📂 RRT_Star│   ├── 📄 README.md│   ├── 📄 rrt_star_algorithm.py│├── 📂 Theta_Star│   ├── 📄 README.md│   ├── 📄 theta_star_algorithm.py│├── 📂 media│   ├── 📹 astar.gif│   ├── 📹 astar_bidirectional.gif│   ├── 📹 dstar.gif│   ├── 📹 dqn_astar.gif│   ├── 📹 rrtstar.gif│   ├── 📹 thetastar.gif│├── 📂 test│   ├── 📄 test_astar.py│   ├── 📄 test_dstar.py│   ├── 📄 test_hybrid_dqn_a_star.py│   ├── 📄 test_rrt_star.py│   ├── 📄 test_theta_star.py│├── 📄 .gitignore├── 📄 LICENSE├── 📄 README.md├── 📄 requirements.txt├── 📄 setup.py├── 📂 .github

To run the algorithms, install the necessary dependencies:

pip install -r requirements.txtpip install -e.

Each algorithm can be executed independently. Example usage:

python A_Star/a_star_algorithm.py            orpytest -v --verbose

Modify the scripts as needed to test different environments or configurations.

The project contains several advanced path planning techniques, including:

A classic grid-based path planning method that computes the optimal path using an 8-connected graph and a Euclidean distance heuristic.

An enhanced version of A* that simultaneously searches from the start and goal, potentially reducing the search time.

A dynamic path planning algorithm that efficiently updates the optimal path in response to changes in the environment.

A rapidly exploring random tree (RRT) is an algorithm designed to efficiently search nonconvex, high-dimensional spaces by randomly building a space-filling tree.

An any-angle variant of A* that uses line-of-sight checks to "shortcut" unnecessary nodes, producing smoother and more direct paths.

An implementation leveraging deep reinforcement learning (DQN) to learn an optimal navigation policy.

Pathfinding results can be visualized using the generated.gif files in themedia directory.

This table provides a performance comparison of various path-planning algorithms implemented in this project.

This comparison helps in selecting the appropriate algorithm based on the application requirements, balancing optimality, efficiency, and adaptability.

• 🟢Green: Best performance in a category.
• 🟡Yellow: Moderate performance.
• 🔴Red: Lower performance or limitations.

• ⚡Fastest Algorithms:
  • DQN-A* (0.01s) andTheta* (0.08 - 0.12s) are the fastest.
• 🛤️Shortest Path:
  • Theta* achieves the shortest path length (82.27) with minimal direction changes (4).
• 🔄Best for Dynamic Environments:
  • Bidirectional A* andRRT* excel in adaptability to dynamic environments.
• 🤖AI-Enhanced Performance:
  • DQN-A* combines high optimality with very high computational efficiency and scalability.

Contributions are welcome! To contribute:

1. Fork the repository
2. Create a new branch (git checkout -b feature-branch)
3. Commit changes (git commit -m "Add new feature")
4. Push to the branch (git push origin feature-branch)
5. Open a Pull Request

If you find this project useful, please consider:

• Starring the repository ⭐️
• Forking the project to contribute enhancements
• Following for updates on future improvements

Your engagement helps increase visibility and encourages further collaboration!

This project is licensed under theMIT License - see the LICENSE file for details.

If you find this repository useful for your work, please consider citing it in your research or projects. Below are the citation details inAPA 7th edition andBibTeX formats.

Gouda Shanbog, D. N. (2025). Path-Planning-for-Intelligent-Mobile-Robots (1.0.1). Zenodo. https://doi.org/10.5281/zenodo.14954689

• BibTeX

@software{gouda_shanbog_2025_14954689,  author       = {Gouda Shanbog, Dada Nanjesha},  title        = {Path-Planning-for-Intelligent-Mobile-Robots},  month        = feb,  year         = 2025,  publisher    = {Zenodo},  version      = {1.0.1},  doi          = {10.5281/zenodo.14954689},  url          = {https://doi.org/10.5281/zenodo.14954689},}

For inquiries or collaborations, reach out viaGitHub Issues.

🎉Happy Coding! 🚀