IoT Machine Learning Based Parking Management System with Anticipated Prediction of Available Parking Spots
IoT Machine Learning Based Parking Management System with Anticipated Prediction of Available Parking Spots
复制标题
基于物联网机器学习的停车管理系统，可预测可用停车位
DOI：
10.1007/978-3-030-97652-1_41
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Lazeroff, Michael
中科院分区：
文献类型：
--
作者：
Chmaj, Grzegorz;Lazeroff, Michael
基于机器学习的设计为识别模式和进行各种预测提供了广泛的手段。在本文中，我们将机器学习应用到物联网架构中，以优化智能停车的访问。我们假设系统会检测到哪个司机把车停在了哪个地点，也会识别出司机回到车里的习惯。这样，我们预测，司机走向停车场或车库的位置很快就会为其他司机所用。寻找停车位的司机将提前收到这样的信息，形式为“车位N将在X分钟内可用”。该设计基于智能手机设备提供的功能：确定停车点，确定步行司机的位置，并作为移动应用程序的基础，因此该系统使用方便，不需要额外的基础设施。
Machine Learning based designs provide an extensive means to recognize patterns and do various kinds of predictions. In this paper we apply machine learning to the Internet of Things architecture to optimize access to smart parking. We assume that the system will detect which driver parked at which spot, and also will recognize driver’s habit of returning to the car. This way we predict that the spot of the driver walking towards the parking lot or garage will soon be available for other drivers. Drivers looking for a parking spot will receive such information in advance, in a form of “Spot N will be available in X minutes”. The design operates over the features offered by smartphone devices: to determine the parking spot, determine the walking driver’s position and also serving as a base for a mobile application, so the system is convenient to use and doesn’t require additional infrastructure.