localization_optimization_using_machine_learning
indoor/outdoor classification. GPS positioning is generally more accurate when outdoors, but there is no GPS information indoors, therefore, other methods should be taken, such as Wi-Fi/ Bluetooth positioning, or base station positioning. However, for the gray areas between indoor and outdoor, many indoor scenes will also have GPS information. At this time, indoor/outdoor classification is required, because indoor and outdoor scenes mean different strategies applied for positioning and navigation, accurate indoor/outdoor classification could provide better positioning and navigation performance and user experience. This project is to extract GPS features from big data, and use machine learning algorithms, then train models to classify indoor and outdoor situations.
