Abstract
The use of automated and unmanned package delivery systems is expanding in both the air and ground domains. Developing a separate infrastructure that can control them both is challenging and cost-prohibitive. This paper investigates the possibility of using established cellular networks to act as a pseudo-independent command and control system linked to a central logic controller to handle the decision-making process for the mobile units. Cellular infrastructure is robust, has multiple backups, and has well-established networks across the nation. The use of these networks can be used to develop air corridors for flying delivery drones, and they can transmit a boundary signal for geofencing to prevent entry into prohibited areas such as airports or military installations without having to program the flying units. The use of edge computing centers is investigated to reduce the latency inherent in centralized networks as distance increases. The limitations of the system related to both air and ground-based unmanned platforms are investigated with recommendations made for the implementation and equipment standardization.
Summary
This investigation researches the viability of using mobile communication, or cellular networks, to act as command and control of automated drone delivery services. The use of mobile networks allows faster connection and more advanced logic used through the system instead of having to build it into the drone (Le Duc, 2019). As the drone travels its assigned route communication latency increases, the use of edge computing centers coupled with the mobile networks reduces latency significantly, and increases reliability. This proposed system is similar to flight following in air traffic control scenarios, the separation between units in the air can be maintained, and routes adhered to (Curtis-Brown, 2017). The system can also be used to control the ground-based units to ensure they maintain their prescribed courses and provide a communication link to the dispatch centers. Though not as critical for ground units as it is for airborne vehicles, edge centers can handle the decision making for traffic situations and provide routing information such as vehicle congestion, path planning, and assist with object avoidance. The research identifies current technologies that are suitable for installation in the infrastructure while proposing standards that work across different platforms that can be adopted by manufacturers. Regulatory solutions are also investigated, particularly for aerial units, as they must operate in the same environment as manned aircraft and could pose a hazard to aviation with the establishment of corridors and ceilings.
Problem Statement
The use of delivery drones, although currently restricted, are beginning to proliferate, the technology is still maturing. The problem is how to adopt a control scheme without significant additional infrastructure and to utilize that which is already in place, which can present difficulties for an emerging technology that operates in both the air and ground environment. The farther away the unit is from the control center causes an increase in latency and a decrease of controllability (Chiang, 2019). Areas within a certain radius of airports must also be restricted; large cities need to have designated air corridors for operation near sensitive areas such as power plants or refineries. It is to be accomplished with current technology to keep costs down and reduce development time. The system must be capable of issuing commands and controlling any drone which can utilize it; therefore, there may be special equipment that is installed on the unit to accept the commands (Curtis-Brown, 2017). The system must be secure against unauthorized access to maintain control and to protect the shipments. This controller must not significantly impair the drone’s ability to complete the assigned task. Ground systems do not experience a weight penalty; the airborne systems require miniaturized and power conserving equipment. Command and control must use the most energy-efficient and longest-range method available to maintain communication while preserving the power supply (Pantelimon, 2019).
Significance of the Problem
The governing authorities are concerned about automated delivery systems and how they interact with the public (Curtis-Brown, 2017). Regulations typically lag technological advances, particularly with emerging industries such as unmanned systems. Although the technology and concepts have been around for at least two decades, the cost has come down significantly to allow the proliferation of different platforms and companies producing them. Standardization of operating procedures must be established and implemented to ensure safe operation while fostering advances (Management, A. I. R. (Ed), 2019). Drone delivery services are beginning to come into service, although currently, only with large corporations, it is not long before startups enter the fray and begin to offer services. These smaller organizations do not have the capital to invest in new infrastructure; therefore, regulations and communication equipment should already be in place and maintained by a governing agency. The Federal Aviation Administration (FAA) is already making strides with its regulations concerning flying drones. Still, they do not have authority over ground operations, but they could fall under the auspices of the Department of Transportation (DOT) (Shen, 2017). Until an agency is designated to handle the regulatory responsibilities, interagency cooperation must suffice until autonomous systems become governed under the same authority.
Alternatives
There are not many alternatives currently available due to a lack of infrastructure and regulations governing unmanned systems, as only a few of these platforms are entering widespread testing (Al-Turjman, 2019). The delivery units are given the destination and are in contact with the control center either through available wireless networks or cellular modems. Not all wireless networks are open access, and the delays associated with accessing them can increase latency in the decision process. There has been some research in Sweden with smart cities and edge centers that is related to the proposed system; however, development is still in the implementation stages; it only concerns autonomous vehicles (Le Duc, 2019).
Cellular Network Antennas
The use of cellular network antennas as waypoints and control centers allows already established infrastructure used in the implementation of the control scheme. Although not as critical for ground units as they will be using established roadways, cellular network antennas allow rapid communication with minimal latency (Flerchinger, 2016). Cellular modems have become small enough to allow the inclusion of this equipment on any platform with minimal design changes (Palash, 2017). The antenna structures, situated at the center of each cell operated by the network, allows for quick handoff between control centers while allowing the establishment of routes that update according to traffic conditions in both the air and ground. As the antenna positions are already known and mapped, using them as available waypoints, air vehicles can be routed along their paths until the destination is approaching and then make the final course adjustments (Chiang, 2019). Once the package is delivered, they will again enter the traffic pattern for control by the network and routed back to the origin point or another hub in the system according to the needs of the operator. This hub system will allow the use of multiple locations and the reallocation of assets as the need arises and for recharging or battery swapping by personnel at the hub for faster turnaround times (Syd Ali, 2019). With the use of machine learning, the assets can be staged based on historical load factors to accommodate the expected orders.
Control Scheme
Once a unit is designated to a package, it will depart the facility and travel along the assigned route. Ground units will utilize the established road infrastructure and will be able to have maps already onboard its system. Since they do not travel as fast as aerial units, they only have to use the onboard logic to deal with traffic and road conditions to reach the delivery point. If there is a significant delay, in response time, established by the operator, then edge computing systems will take over the operation of the unit until latency is within limits, and stability is restored (Le Duc, 2019). The edge systems will be placed at the antenna sites to allow quick acquisition of control and act as the traffic manager. Each edge system will have decision authority over the cell in which it operates. Airborne units are controlled in the same manner; still, a higher level of authority is used as they cannot have as much onboard logic and redundancy as ground units because of weight restrictions.
Unmanned Ground Vehicles
Unmanned ground vehicles (UGV) will be able to utilize the same systems as the flying units but in a different manner. UGVs will typically have much longer endurance than the airborne units so they can be loaded with more parcels and allocated to deliveries that are not time-sensitive (Trösterer, 2017). They are utilized for longer distances based on the vehicle type; this can even be implemented into long haul autonomous trucks as the cellular infrastructure is typically along most of the major highways in the country. Currently, these autonomous trucks require a safety driver in case of problems, with the use of edge centers and the cellular networks, control can be continuous, the destination and route updated depending on conditions (Trösterer, 2017). The unit could even be directed to refuel or report maintenance issues as they happen. The operators can then asses the problem and decide to continue to the destination or interrupt the trip to perform the repairs. The proposed system can also be used as a combination of both air and ground units with a hub and spoke system. The ground units can be utilized to deliver from the central warehouse to a localized distribution center that is segregated by zip codes to other logistic centers. Then the flying units are employed for the last mile, depending on the size of package and weight restrictions.
Unmanned Aerial Vehicles
The use of unmanned aerial vehicles (UAV) will require a higher degree of control and lower latency than ground units. Although the control logic has improved significantly and algorithms refined due to miniaturization and other advances in technology, the environment in which they operate requires faster response times (Motlagh, 2016). The decision times are significantly shorter; it is much safer to exert continuous control over the unit and have established air corridors in which they operate with ceilings and geofences in place to prevent interference with manned aircraft (Pan, 2019). Using the edge centers to their full potential, the use of the onboard cellular modem will allow direct control over the UAVs within the sphere of influence of the particular cellular site concerning routing, speed, and separation in the air. Instead of having the onboard cellular modem connected to the central controller, the unit will be connected to edge centers along its path and then handed off to the next as the journey progresses (Gharibi, 2016).
The edge computing components reduce latency in the decision process and alleviates the central controller, only the destination, original route, and primary control shortly after leaving the distribution center require action by the origination controller (Le Duc, 2019). Once the unit leaves the departure point cell, control is handed over to the edge center, which is in communication with the unit through its modem and tracked by the automatic dependent surveillance broadcast (ADS-B) equipment mounted on the unit. The ADS-B will receive information from the flight controller and onboard logic to coordinate with the edge center for route and altitude instructions. ADS-B will also ensure tracking by the governing authority over the air space and that they remain in established air corridors along the route and away from restricted air space (Curtis-Brown, 2017).
Edge Computing
Edge systems function at the outer periphery of the internet and provide decreased latency while maintaining a high level of control over the associated units. Instead of having to check in with the centralized controller continually, the unit is controlled under the local node within the same cell as the antenna it is in communication with. This control will allow the unit to proceed at the highest efficiency to its destination to minimize downtime and to increase the utilization of the asset. The use of a centralized control scheme alleviates the designer from having complicated control logic onboard the vehicle and allows the use of more streamlined algorithms. A majority of the processing is done off of the machine (An, 2019).
The control of the unit will function much like flight following does by the FAA for small aircraft. Once the vehicle enters the sphere of control for a particular node, either ground or air-based, then that unit will receive and follow the instruction from that node until it passes to the next with pre-arranged handoffs based on routing and destination (Syd Ali, 2019). For example, ground units will be routed based on the available traffic conditions and roads to the destination. The edge center will be in contact with the unit the entire time for navigation and object avoidance. In case of emergencies, the onboard logic will have the ability to take action to prevent damage to itself and other vehicles (Zang, 2019). The ground vehicles will be capable of complete autonomous operation as they will not be limited by the weight restrictions that flying units will be, and their systems and algorithms can be more complicated and redundant without the weight penalties that airborne units experience.
The edge center themselves are installed into the same site as the cellular antenna; this allows the system to take advantage of already established infrastructure while reducing costs and maintenance associated with the introduction of such a system (Le Duc, 2019). The antenna infrastructure typically has redundant power supplies and back up batteries that are used in case of emergencies; this adds additional layers of safety to the system to prevent power interruption. The edge centers are then connected to the central controller that will monitor the assets and allow the operators to monitor the health of the vehicle and to issue new instructions if necessary, such as a hold on operations based on weather events.
Regulatory Concerns
Currently, the FAA is responsible for the regulation of drones concerning air operations under part 107 of the code of federal regulations (CFR) (Pan, 2019). Ground-based vehicles are under the auspices of the Department of Transportation (DOT), which coincidentally is the parent organization of the FAA. As these two agencies are directly related, and as the use of autonomous drones proliferates, there will be overlaps in the regulations concerning these systems. As of yet, no single agency has control over autonomous transportation. Still, as they begin to become integral to the operations of logistics companies and other businesses, there will need to be laws established and a governing agency that has sole responsibility for these systems. For example, pilots of drones must be certified if it is involved in commercial uses or for hire, drones over 55 pounds are also considered aircraft by the FAA and must be registered as such (Clothier, 2015). Small aerial delivery drones may not approach this weight class because the larger packages will be delivered by the ground units, as these systems will be autonomous, they will need to be certified by the FAA and tested thoroughly. A control software will need to be developed that is approved by the regulatory agency; it will most likely have to be open-source since multiple platforms and manufacturers will be using the same infrastructure. Each unit will have to have a unique identifier for licensing and tracking purposes; airborne units will have to have ADS-B installed or similar equipment so that the controller software can monitor them in flight and the units can maintain proper separation and altitudes along their route (Shen, 2017). UGVs will have to be treated similarly to the UAVs, software, and control schemes will need to be certified by the governing authority. The ground-based systems will be interacting with humans on a higher level. They will have to demonstrate the ability to deal with the dynamic environment and the actions of unpredictable humans.
In 2018 the DOT was directed by the president to begin research into developing a framework for the demonstration and approval of automated vehicles (U.S. Department of Transportation, n.d.). The problem is an overlapping regulatory authority that may not agree between agencies that cover the same aspects of the operation. A single regulating body must be developed to address the autonomous services in the transportation network of the country. It would allow the streamlining of the regulations and rapid responses to the development and implementation of new and emerging technologies while ensuring that the safety issues are addressed promptly and that the infrastructure is designed and maintained per established regulations for the safe operations of the system.
Limitations
The proposed system is not without its limitations, as the cellular antennas are privately owned property, and rights must be secured to install and maintain the equipment required to implement the system. Not every antenna owner will be interested in participating in the program. Some antennas are atop high tension power lines, so these are excluded from the established air corridors for safety reasons to prevent contact in case of loss of control. Air corridors will need to be established to avoid the flying units from interfering with larger aircraft. Equipment will need to be developed that can transmit a boundary signal in the vicinity of these areas to instruct the logic onboard the drones from entering the area to act as an additional layer of geofencing instead of relying solely on map data for which the logic may not be able to store onboard due to hardware restrictions (Bae, 2019).
The system will also need to be secured against spoofing to prevent rerouting of the units to avoid the theft of the deliveries, either using an advanced but lightweight cipher or a blockchain method (Garcia-Magarino, 2019). There will also be required a backup communication method in case of loss of signal by the cellular modem. Developments and advances in technology have brought about the miniaturization of wireless ethernet capabilities that make them lightweight and energy-efficient, this equipment could be used to connect to available wireless networks for both navigation and instructions (Flerchinger, 2016). Secondary communication protocols will have increased latency as the range is limited by these systems, and access to the wireless networks will have to be negotiated in advance by the providers.
UGVs will be subject to similar command and control limitations. Depending on their size, their role may be used to shuttle deliveries between hubs with the final mile being done by the airborne units if the cargo is light enough, larger packages may still require human assistance to unload or arrangements made with the recipient. Ground-based units will have increased latency because of the line of sight issues with the topography in urban settings (Dorling, 2016). While traveling through an area with tall buildings, communication could be disrupted, and the onboard logic will have to assume control until reestablished. The size of the vehicle must also be considered as the units may be designed to run on the sidewalks as well as the regular roads, and the logic must be able to deal with pedestrians and manned vehicles.
Equipment
The size and purpose of the vehicle is directly proportional to the amount of logic, computing power, and communication equipment it can carry aboard. Airborne units are particularly susceptible to weight issues, for maximum efficiency and load carrying, they must be as light as possible while still having the equipment onboard to control them effectively. The ADS-B will have to be integrated into the system while not only reporting information to the operators, will also have to report to the FAA as well. (Syd Ali, 2019). Miniaturization has brought about a unit that can be equipped to a drone; manufacturer DJI is equipping all of their drones over 250 grams with one from the factory (Lillian, 2019). Which will, of course, require the server nodes at the cellular sites to be able to receive and decode the information transmitted by the ADS-B system for the purpose of traffic control. The ADS-B equipment will not be the sole source of information for control nodes since cellular towers will be used; they will also have modems that operate in multiband frequencies to allow the use of different bands along the route and across carriers (Abdelhamid, 2015). The type of reporting by both the ground and air units can vary, as the air units will travel much faster, but the cellular modems can be used to verify the ADS-B transmissions and to receive the instructions (Lin, 2019). It can be as simple as text exchanges with data from the navigation system and flight controller to low bandwidth video to aid the human pilot if they need to manually take control of the system (Pantelimon, 2019).
Cellular modems vary in design by manufacturer, recent advances have brought about the development of stand-alone boards capable of communicating on all bands with very low power consumption (Flerchinger, 2016). This lightweight solution increases the communication and control capabilities of the units without adding complexity to the system as the infrastructure is already well established. As the proposed method will typically be in constant communication with the node in one manner or another and will be able to switch when the need arises. The modem, combined with wireless internet technologies, allows decreased latency for teleoperation when required by human controllers (Al-Turjman, 2019). Ground units that are used for delivery will not be required to maintain constant contact as they will be operating on the established roadways and will only need to check in to receive updated routing information or reassignment after the delivery is made. The onboard logic will be able to handle most of the decisions, only when the latency passes a certain threshold set by the operator will the edge center take over until communication protocols are within limits.
Communication Security
Any signal that is transmitted can be intercepted or jammed. Because the proposed system relies heavily on wireless transmissions, a security protocol must be established to prevent interference with operations (Bae, 2019). The problems are, adopting a protocol that multiple companies and manufacturers can use and agree upon as the system could be considered a utility. The use of blockchains, developed for digital currency, is the most probable solution as the edge centers can verify them along their pathway and prevent tampering with the instructions as each command will have its verifiable fingerprint attached to it (Liang, 2017). Encrypting of the communications will also be required, again a cipher protocol will have to be agreed upon ahead of time, generated by the control network and assigned to each drone to prevent take over from those looking to obtain the cargo or use the drone for unlawful purposes (Garcia-Magarino, 2019). Outright jamming may be possible, but not without disrupting the entire network, which will alert the control centers that the signal is interrupted.
If the ground units are using open wireless networks, a virtual private network is required, although wireless networks do not offer the range of cellular modems, they could act as a backup system to verify connection if the primary means of communication is lost or disrupted (Pantelimon, 2019). To further enforce the security of the system, the units, both air and ground, can act as a peer to peer network within the control node checking the blockchain identification as a backup system to the edge centers, This would allow the edge center to use the onboard resources of the drones to expand its capabilities with the use of cloud computing while the drone is within its sphere of control (Garcia-Magarino, 2019). If communication is lost, a contingency plan with preprogrammed coordinates can be enacted to direct the drone to the nearest service center or distribution point to have the cargo swapped to another vehicle for the continuation of the delivery. The security of the system may have to be enforced by regulation of the governing authority or a combination of agencies, such as radios in aircraft that operate under the rules of the Federal Communications Commission. As unmanned deliveries are a new segment of the autonomous industry, new methods of securing the communication protocols are to be developed rapidly using existing technology to prevent the unauthorized use of the platforms and to protect the assets and people that may be in the vicinity.
Recommendations
Autonomous delivery systems are an emerging technology that has the potential to reshape logistics; this requires the infrastructure to be in place and open source technology that is available for any manufacturer to implement into the vehicle architecture. It must be lightweight, the operating system must be secure, such as a Linux based that is open-source and customizable, and multiple manufacturers of autonomous systems must adopt the equipment. Ground-based systems can use the established roadways and have enough logic on the vehicle to handle most situations it encounters. Still, airborne systems require a low latency established control system; this is accomplished through the addition of edge computing centers into the cellular network at strategic antennas to establish flight corridors and to maintain constant communication with the operator. Each cell in the antenna system can act as a pseudo air traffic control to maintain separation, course, and speed of the drones and updating delivery instructions. Once the drone is within a specified distance of the destination, then the onboard logic assumes control for the final segment of the operation. Which increases the reliability of the system and adds an extra layer of safety in case of malfunctions in both the airborne and ground-based delivery systems as terrestrial systems still require low latency communication with the operator to ensure delivery.
These systems must be regulated in the same manner as a utility so that there is compatibility across platforms, and they can use the same infrastructure without significant changes and upgrades. Statutory issues for unmanned systems should be brought under the auspices of a single governing authority, and this can be problematic as they operate in different domains. It may be advantageous to have a division within each governing body that is responsible for the regulations concerning autonomous systems. Still, they must also be concerned with how these systems interact with each other.
Conclusions
The use of autonomous vehicles for package delivery is inevitable, and new technologies are to be developed and integrated into existing systems so that different companies can use them. This method of control is similar to how the FAA routes aircraft and controls departures, used by various airlines all under the same set of rules and procedures. An established framework of regulations and infrastructure can be developed to accommodate this emerging industry. It is the method of command and control that poses the most significant challenge; the technology exists today to integrate the systems into current infrastructure. Still, it is unknown as to who would control it. Would the system be a public-private enterprise, or would a private company install its edge nodes for control of the delivery systems?
Integrating this system into the cellular communication network will take the cooperation of the antenna owners and the leaseholders. The use of the edge centers will have to report information to the FAA concerning flying units to ensure they do not interfere with air traffic, and the FAA will have final authority over their flight plans. The DOT must mandate separate rules for autonomous vehicles that operate on the roads and highways, and safety must be the primary concern with the delivery of the cargo secondary. Autonomous delivery systems can be a prolific yet disruptive technology that must be highly regulated as it will operate along with manned vehicles and aircraft. The technology currently exists to implement this strategy, and it only requires the modification of current infrastructure to accomplish it and the implementation of rules that must be enforced by a governmental agency.
References
Abdelhamid, A. Z. (2015). Advanced software ground station and uav development for nlos control using mobile communications. Discrete Dynamics in Nature and Society. doi:http://dx.doi.org.ezproxy.libproxy.db.erau.edu/10.1155/2015/285746
Al-Turjman, F. (2019). Drones in iot-enabled Spaces. Boca Raton, FLorida: CRC Press. Retrieved from https://ebookcentral.proquest.com/lib/erau/detail.action?docID=5813363.
An, N. Y. (2019). Seamless virtualized controller migration for drone applications. IEEE Internet Computing, 51-58. doi:10.1109/MIC.2018.2884670
Bae, M. K. (2019). Authentication and delegation for operating a multi-drone system. Sensors, 19(9). doi:http://dx.doi.org.ezproxy.libproxy.db.erau.edu/10.3390/s19092066
Chiang, W. C. (2019). Impact of drone delivery on sustainability and cost: realizing the uav potential through vehicle routing optimization. Applied Energy, 1164-1175. doi: https: // doi.org/10.1016/j.apenergy.2019.03.117
Clothier, R. A. (2015). Structuring the safety case for unmanned aircraft system operations in non-segregated airspace. Safety Science, 213-228. doi: https: //doi.org /10.1016 /j.ssci. 2015.06.007
Curtis-Brown, N. G. (2017). UAV collision cetection and cvoidance using ads-b Sensor and custom ads-b Like solution. AIAA Information Systems. Grapevine, TX. doi:https://doi-org.ezproxy.libproxy.db.erau.edu/10.2514/6.2017-1153
Dorling, K. H. (2016). Vehicle routing problems for drone delivery. IEEE Transactions on Systems, Man, and Cybernetics: Systems. doi:10.1109/TSMC.2016.2582745
Flerchinger, B. F.-P. (2016). Field testing of 3g cellular and wireless serial radio communications for smart grid applications. 2016 IEEE Rural Electric Power Conference. Westminster, CO: IEEE. doi:10.1109/REPC.2016.22
Garcia-Magarino, I. L. (2019). Security in networks of unmanned aerial vehicles for surveillance with an agent-based approach inspired by the principles of blockchain. Ad Hoc Networks, 72-82. doi:https://doi.org/10.1016/j.adhoc.2018.11.010
Gharibi, M. B. (2016). Internet of Drones. IEEE Acess, 1148-1162. doi:10.1109/ACCESS.2016.2537208
Le Duc, T. L. (2019). Machine learning methods for reliable resource provisioning in edge-cloud computing: a survey. ACM Computing Surveys. doi:10.1145/3341145
Liang, X. Z. (2017). Towards data assurance and resilience in iot using blockchain. MILCOM 2017 – 2017 IEEE Military Communications Conference. Baltimore, MD: IEEE. doi: 10.1109/MILCOM.2017.8170858
Lillian, B. (2019, May 22). DJI to add ads-b receivers to new drones. Retrieved from Unmanned Aerial Online: https://unmanned-aerial.com/dji-to-add-ads-b-receivers-to-new-drones
Lin, C. H. (2019). An ads-b like communication for utm. 2019 Integrated Communications, Navigation and Surveillance Conference (ICNS). Herndon, VA: IEEE. doi: 10.1109/ ICNSURV.2019.8735350
Management, A. I. R. (Ed). (2019). Unmanned aerieal vehicles: breakthroughs in research and practice. Hershey, Pennsylvania: IGI Global. Retrieved from https: //ebookcentral .proquest.com/lib/erau/detail.action?docID=5746432.
Motlagh, N. H. (2016). Low-altitude unmanned aerial vehicles-based internet of things services: comprehensive survey and future perspectives. IEEE Internet of Things Journal, 899-922. doi:10.1109/JIOT.2016.2612119
Palash, R. D. (2017). A high power fully integrated single‐chip CMOS transmitter for wireless communication of unmanned aircraft system. Microwave and Optical Technology Letters, 432-439. doi:https://doi-org.ezproxy.libproxy.db.erau.edu/10.1002/mop.30305
Pan, Y. L. (2019). When uavs coexist with manned airplanes: large‐scale aerial network management using ads-b. Transactions on Emerging Telecommunications Technology, 30(9). doi:https://doi-org.ezproxy.libproxy.db.erau.edu/10.1002/ett.3714
Pantelimon, G. T. (2019). Survey of multi-agent communication strategies for information exchange and mission control of drone deployments. Journal of Intelligent & Robotic Systems, 779-788. doi:https://doi-org.ezproxy.libproxy.db.erau.edu/10.1007/s10846-018-0812-x
Shen, Z. C. (2017). A dynamic airspace planning framework with ads-b tracks for manned and unmanned aircraft at low-altitude sharing airspace. 2017 IEEE/AIAA 36th Digital Avionics Systems Conference (DASC). St. Petersburg, Florida: IEEE Xplore. doi:10.1109/DASC.2017.8102125
Syd Ali, B. (2019). Traffic management for drones flying in the city. International Journal of Critical Infrastructure Protection. doi:https://doi.org/10.1016/j.ijcip.2019.100310
Trösterer, S. M. (2017). Transport companies, truck drivers, and the notion of semi-autonomous trucks: a contextual examination. 9th International Conference on Automotive User Interfaces and Interactive Vehicular Applications (pp. 201-205). Oldenburg, Germany: Association for Computing Machinery. doi:10.1145/3131726.3131748
U.S. Department of Transportation. (n.d.). USDOT automated vehicles activities. Retrieved from U.S. Department of Transportation: https://www.transportation.gov/AV
Zang, H. Z. (2019). Object tracking for a smart city using iot and edge computing. Sensors, 19(9). doi:http://dx.doi.org.ezproxy.libproxy.db.erau.edu/10.3390/s19091987
Patrick Carroll 