关键词: 大数据分析, 大数据架构, 无人机云数据, 无人机运行分析, 中国民用航空局

Abstract:

After several years of development, light and small unmanned aircraft systems (UASs) have been widely used in various industries both in China and many other countries. However, the UASs have many models, with scattered equipments and no systematic management. Meanwhile, some safety issues exist. This urgently requires the relevant regulatory authorities to regulate, supervise, and maintain safe flight operations by taking their operating rules and characteristics into account. In order to standardize the operation of light and small civil UASs across the country and promote the industry development, the Civil Aviation Administration of China has issued the "Provisions for the Operation of Light and Small Unmanned Aircraft (for Trial Implementation)" advisory circular and the "Specification for Interface Data of Unmanned Aircraft System Cloud System". The UAS cloud data exchange platform was developed in 2016, and the data sharing of multiple UAS cloud systems in China was realized. With this platform the UASs registered in different UAS cloud systems are visible to each other in the same airspace, which improved the flight safety of China's low-altitude airspace. However, with the rapid development of the application of the unmanned aerial vehicle (UAV) industry, the number of UAVs supervised by the Civil Aviation Authority and the data on the operation of the UAVs has increased dramatically, which has also brought great challenges to the traditional data management methods. In this paper, we will describe the current situation of the operation of big data from the UAS cloud data exchange platform in China followed by discussion on the technical bottle necks in the statistical analysis of the operation data of the traditional UAS. Then we will propose a statistical analysis method for the UAS operation data, and establish a framework of statistical analysis of big data from the cloud data exchange platform. In the end, we will outline how to use Apache Spark and Cassandra database to quickly process, store, count, and analyze the massive data generated by the UAS cloud data exchange platform. The research situation of implementing various statistical index algorithms based on the platform is introduced in detail. This research not only improves the efficiency of statistical analysis of UAS operation data, but also provides the operation management rules of China's light and small UASs from multiple dimensions. sWe highlight that the UAS has significant operational characteristics, which are different from general and transportation aviations. This paper provides reference for government and industry decision-making, which has strong practical significance.

Key words: big data analysis, big data architecture, UAS cloud data exchange, UAS operation analysis, Civil Aviation Administration of China