Path Planning Solution for Intelligent Robots Using Ex*RRT Algorithm in Disaster Relief

Unmanned Ground Robots (UGVs) are increasingly becoming vital tools in disaster relief operations, offering a means to navigate through hazardous environments where human intervention may be risky or impossible. These intelligent robots are designed to perform a variety of tasks, such as searching for survivors, delivering supplies, and assessing damage. However, the effectiveness of UGVs in such critical missions largely depends on their ability to navigate complex and unpredictable terrains, a challenge that necessitates advanced path-planning strategies. The paper introduces a novel approach to path planning for UGVs in disaster relief scenarios. This approach combines the strengths of advanced rapidly Rapidly-exploring Random Trees (RRT) algorithms with the capabilities of YOLOv4, a state-of-the-art technique for object detection, specifically tailored here for human detection. RRT algorithms are known for their efficiency in exploring complex spaces, making them ideal for navigating through the unpredictable terrains typical of disaster zones. Meanwhile, YOLOv4 enhances the UGVs' ability to detect and respond to human presence, a critical requirement in rescue operations. The UGVs equipped with this novel path-planning strategy were able to navigate safely and efficiently, adapting to the dynamic nature of the environment. More importantly, the accuracy in detecting human presence was significantly improved, a crucial factor in ensuring that aid and rescue efforts are directed where they are most needed. By enhancing the path planning capabilities of UGVs with advanced algorithms and human detection technology, this approach holds the potential to revolutionize disaster response operations, making them more efficient and impactful, ultimately leading to faster and more effective aid delivery to those in dire need.