Improving Path Efficiency and Adaptability in Robot Path Planning with I-A

This study introduces I-A*, an improved path-planning algorithm designed to enhance the classical A* approach by increasing computational efficiency, shortening path length, and reducing waypoint counts. Addressing the limitations of traditional Data-Driven Learning Based (DDLB), Memory-Based (MB), and Sample-Based (SB) algorithms, I-A* integrates safety check techniques to adjust to environmental changes. Extensive simulations on various maps with diverse obstacle densities and configurations reveal that I-A* outperforms both the standard A* and Rapidly-exploring Random Trees (RRT) in computational speed, path accuracy, and efficiency. Results indicate that I-A* reliably produces shorter and more efficient paths with fewer waypoints, making it suitable for real-time navigation applications in robotics and autonomous vehicles.