An Indoor Navigation System for Autonomous Shopping Cart

Indoor autonomous robots are increasingly being applied in the retail sector to assist customers in shopping more efficiently. This paper describes an autonomous robot for a shopping assistant which finds an optimal total path based on the user's shopping list. The system uses K-means++ to cluster goods locations, A* to determine the shortest path between items, and Genetic Algorithm (GA) to find the globally optimal route. Mapping, localization, and navigation algorithms are combined in this system to ensure smooth movement within the store environment. All algorithms use a centralized database, which reduces computational load for each algorithm. The system is implemented on a real robot in a test environment similar to a store, demonstrating high performance and strong potential for practical deployment.