Design and Develop Optimal Pathfinding Algorithm Applied in Book Accessing and Returning for Autonomous Mobile Robot in Library

This paper presents a visual optimal pathfinding algorithm that avoids obstacles on the way to the destination in a simulation environment, and the idea took inspiration from the optimization for solving the Traveling Salesman Problem. Firstly, using a webcam to scan the book information by QR code to know where the book's location is, what is the type of the books and use location information to send to robots for autonomous navigation duty. After that, the order list of books borrowed or returned is calculated and visualized using the K-means Clustering and Genetic algorithm. Then, to apply the optimal algorithms for a practical situation, Robot Operating System (ROS) is considered. In the ROS platform, to make sure the robots operate precisely, parameters of global, local cost maps and algorithms for global, local planners need to be adjusted. The robots will follow the order calculated path and announce when arriving at the book's location successfully. To track if the robots work correctly, a software for users to observe robot operation is created. The proposed algorithm works quite well in a simulation environment through implementing experiments. It is also a good foundation for further research into another autonomous field.