1 Unmanned Aerial Vehicle enabled Internet of Everything: Opportunities and Challenges Yalin Liu, Hong-Ning Dai, Senior Member, IEEE, Qubeijian Wang, Mahendra K. Shukla, Muhammad Imran Abstract—The recent advances in information and communi- system. In summary, IoE is expected to achieve three expec- cation technology (ICT) have further extended Internet of Things tations: 1) scalability means to establish a scalable network from the sole “things” aspect to the omnipotent role of “intelligent architecture with ubiquitous coverage; 2) intelligence implies connection of things”. Meanwhile, the IoE realization meets critical challenges including the restricted network coverage and to construct a global computing facility enabling intelligent de- the limited resource of existing network technologies. Recently, cisions; 3) diversity indicates supporting diverse applications. Unmanned Aerial Vehicles (UAVs) have attracted significant Therefore, the realization of IoE essentially depends on the attentions attributed to their high mobility, low cost, and flexible achievement of the above three expectations. deployment. Thus, UAVs may potentially overcome the challenges We have witnessed the rapid development of ICT technolo- of IoE. This article presents a comprehensive survey on opportu- nities and challenges of UAV-enabled internet of everything (IoE). gies. In particular, ICT technologies have further extended We first present three critical expectations of IoE: 1) scalability existing human-oriented Internet to machine-oriented Internet requiring a scalable network architecture with ubiquitous cover- of Things, which consists of wireless sensor networks (WSN) age, 2) intelligence requiring a global computing plane enabling for connecting multiple sensor nodes via an self-organized intelligent things, 3) diversity requiring provisions of diverse topology [1], low power wide area network (LPWAN) for applications. Thereafter, we review the enabling technologies to achieve these expectations and discuss four intrinsic constraints offering large-range coverage of power-constrained nodes [2], of IoE (i.e., coverage constraint, battery constraint, computing and 4G and 5G mobile networks for supporting massive-access constraint, and security issues). We then present an overview of services of machine-to-machine (M2M) communications [3]. UAVs. We next discuss the opportunities brought by UAV to IoE. Meanwhile, massive data are generated from various things in Additionally, we introduce a UAV-enabled IoE (Ue-IoE) solution real-time manner. The breakthrough of artificial intelligence by exploiting UAVs’s mobility, in which we show that Ue-IoE can greatly enhance the scalability, intelligence and diversity of IoE. (AI) technologies integrating with massive IoT data brings Finally, we outline the future directions in Ue-IoE. the opportunities in realizing intelligent applications including intelligent recognition, intelligent management and intelligent Index Terms—Unmanned Air Vehicles, Internet of Everything, Internet of Things, Edge Intelligence, Multi-UAV Ad Hoc net- decision [4]. As a result, conventional IoT has been evolved works, Trajectory Optimization into IoE that supports intelligent connection of things, thereby enabling smart applications, such as smart meter [5], smart grid, smart manufacturing [6], smart agriculture, intelligent I. INTRODUCTION traffic scheduling [7]. In this regard, the existing wireless communication networks (including WSN, LPWAN and 5G NTERNET of Everything (IoE) represents a fantastic vision cellular networks) can help to realize the scalability of IoE. everything I in the future, in which is connected to the Meanwhile, the cutting-edge big data analysis technologies Internet, thereby offering intelligent services and facilitating and artificial intelligence can be used for enabling intelligence decision-making. IoE’s implementation depends on interdis- to IoE. Finally, the emerging IoE intelligent services have ciplinary technical innovations such as sensor and embedded proliferated a huge market for diverse IoE applications. technologies, low power communications and big data analyt- However, there are still a number of intrinsic limitations ics. First, the advances in sensor and embedded technologies preventing IoE from achieving the above three expectations. have made the IoT nodes being more portable and less energy In particular, IoE has network coverage/access constraints, bat- consumption. Second, the appearance of Low Power Wide tery constraint of IoE nodes, security and privacy vulnerabil- Area Network (LPWAN) technologies offers the ubiquitous ities. First, the existing network infrastructures are coverage- network connection of low power IoT nodes. Furthermore, constrained in some harsh and remote geographical areas due the breakthrough in artificial intelligence and the availability to the restricted deployment of network infrastructures. Hence of massive IoT data have driven the intelligence of IoE. In this the ubiquitous connections of IoE cannot be achieved. In ad- way, IoE can be applied in wide applications such as smart dition, IoE nodes also suffer from the limited battery capacity manufacturing, smart agriculture and intelligent transportation due to cost and portability considerations. As a result, the battery-constrained nodes are easily exhausted and eventually Y. Liu, H.-N. Dai and Q. Wang are with the Faculty of Information Tech- nology, Macau University of Science and Technology, Macau SAR (email: lead to the connection-lost. This case is especially severe for [email protected]; [email protected]; [email protected]). the nodes in the coverage-constrained areas. Furthermore, most M. K. Shukla is with the Department of Electrical and Computer Engineer- of the nodes in IoE have no enough computing capability to ing, University of Saskatchewan, Canada (email: [email protected]). M. Imran is College of Applied Computer Science, King Saud University, process local sensor data. Moreover, the over-simplified access Riyadh, Saudi Arabia. (email: [email protected]). protocols (e.g., NB-IoT and LPWAN) also pose potential 2 security vulnerabilities in IoE, such as information being enhance scalability, intelligence, and diversity of IoE. Further, eavesdropped or being forged by malicious relay nodes. To we outline crucial issues in Ue-IoE as well as future directions address these challenges, IoE requires a flexible-coverage and in SectionV. Finally, SectionVI concludes the paper. elastic-deployment so as to achieve the ubiquitous coverage and offer quick response in a highly-efficient and reliable way. II. INTERNET OF EVERYTHING In recent decades, Unmanned Aerial Vehicles (UAVs) This section first presents three key expectations of IoE have attracted significant attentions from both industrial and in Section II-A, then outlines the enabling technologies to academia communities due to their flexible and elastic ser- fulfill the expectations in Section II-B, and next discusses the vices. With high mobility and elastic deployment, UAVs can challenges in Section II-C. extend the coverage of IoE [8], [9]. In particular, compared with terrestrial networks and satellite remote communications, A. Three expectations of IoE low-altitude UAVs-enabled wireless networks can be quickly deployed and be flexibly reconfigured to enhance the net- The vision of IoE is to connect ubiquitous electronic devices work coverage and capacity [9]. Furthermore, the flexible (i.e., terminal nodes of IoE) to the Internet, then to analyze deployment of UAVs also enables myriad IoE applications. massive data generated from connected terminal nodes, and Through the dedicated configuration and specified control thereby to offer intelligent applications for the advancement design, UAVs have a variety of applications such as agriculture of human society. To achieve this vision, IoE is expected to management [10], marine mammals monitoring [11], and fulfill three key expectations: 1) scalability means to establish military medical evacuation [12]. However, using UAV in IoE a scalable network architecture with ubiquitous coverage; 2) also poses new challenges in the perspectives of dynamic intelligence implies to construct a global computing facility network connection, flexible network topology, precise control enabling intelligent decisions; 3) diversity indicates to support and lightweight intelligent algorithms. In order to address these diverse applications. Fig.1 shows three expectations as well challenges, we need to design the dedicated communication as their typical enabling technologies. In detail, we describe protocols, the flexible resource allocation mechanism, the the three expectations as follows. optimal trajectory plans of UAVs and the portable intelligent 1) Scalability: Scalability means to establish a scalable net- algorithms in IoE. work for IoE to elastically cover everywhere and every- Although several surveys have already discussed the IoE thing. In this sense, IoE can satisfy various communication vision in recent decades [13]–[16], they mainly focused on a requirements for different geographical scenarios including single IoE technology. For example, the authors in [13] have urban, rural, underwater, terrestrial, aerial, and space. To considered the Internet Protocol version 6 (IPv6) to realize achieve this goal, the scalable IoE network requires wide the ubiquitous communication access to the Internet while coverage, massive access, and ubiquitous connection. Such they ignored the issues about wide coverage