Bi-directional virtual search algorithm for efficient and collision-free path planning in autonomous robots navigating static and dynamic environments

This study proposes an efficient and adaptable path-planning model for autonomous mobile robots operating in both static and dynamic grid-based environments. Initially, a collision-free grid map is constructed to represent the configuration space, ensuring accurate placement of obstacles and target coordinates. The robot navigates toward its goal through a dynamic decision-making algorithm that updates its movements based on real-time coordinate comparisons and environmental changes. Extensive experiments were conducted across multiple static and dynamic scenarios. The proposed model achieved a path efficiency improvement of 17.9 % and a computational time reduction of 23.4 % compared to traditional A* and Dijkstra’s algorithms. The success rate remained consistently above 96.5 % even in environments with moving obstacles. Key evaluation metrics, including path optimality, success rate, average computation time, and adaptability score, demonstrate the model’s superiority over benchmark methods. Overall, the proposed framework ensures robust, real-time path planning with minimal computational overhead, making it highly suitable for complex autonomous navigation tasks in uncertain environments.