The mesh network envisioned uses unmanned aircraft to deploy solar-powered beacons in regions lacking Wi-Fi or cellular coverage. Designing the beacons required analysis of battery life using solar panels, mechanical housing survivability and longevity of electronic components. The team prototyped and tested six beacons, all with minimal size and weight so they could be dropped by drone. A mobile application allows users to send emergency requests, display their location, and receive and display incoming responses from the main station. The main station is responsible for receiving and responding to emergency requests, monitoring the beacons and keeping a log of the system. The beacons are responsible for handling these requests. With no cellular service, the main station communicates with the beacons over the radio mesh network.

The ground-based optical target tracking system is an autonomous tracker designed to be used by two or more persons. The system tracks a full-size target vehicle moving at approximately 28 kilometers per hour at distances ranging from 100 to 300 meters. The system is composed of six principal parts: a camera, a PC, a microcontroller, a motor driver, a stepper motor and a laser pointer. The camera sends target images to the PC for image processing while the target remains in the camera’s field of view. The image-processing algorithm determines changes in target position between two separate images and converts this to an angle that the pointer must move to remain pointing at the target. The microcontroller receives the angle input and converts it to a frequency value to send to the stepper driver –while incorporating error values like time delay into the final frequency. The driver reads this value and produces a frequency pulse to drive the stepper motor. The stepper motor then rotates a turntable that the pointer is mounted to. The pointer remains on the target vehicle until it exits the camera’s field of view.

There is currently no safe method for recovering a small fixed-wing unmanned aircraft while at sea on smaller naval vessels. The team created a system, deployable from small watercraft, that can capture a fixed-wing unmanned aircraft without the use of landing gear. The system is easily stowed and deployed, and able to weather the wide range of operating conditions typical of a maritime environment. The design uses a structural frame with nylon netting to safely arrest drones that fly into the netting.
A frictional braking system attached to the net dissipates momentum to minimize potential damage to the drone. The system also includes lighting that allows unmanned aircraft pilots to detect and align with the capture net during night operations. A three-man crew can erect or stow the system in 15 minutes. The system was tested against International Organization for Standardization and military standards to ensure it met all of its requirements and could survive a wide range of operating environments.

Date trees can grow to over 80 feet tall and, in their natural habitat, have only the wind to depend on for pollination. Traditionally, agricultural workers pollinate date trees using large, hose-like machines, sometimes in conjunction with a lift table or similar machine to get closer to the date flowers.The team developed an unmanned aircraft pollination approach to increase the efficiency of this process. Teams 18054 and 18055 developed the dispenser and ground control systems. This project focuses on the communication between the unmanned aircraft and the operator, using a new “internet of things” product, the Hologram Nova, to give the unmanned aircraft LTE, or long-term evolution, capabilities virtually anywhere in the United States. The team developed a web application to autonomously create paths for each unmanned aircraft, while giving users the ability to control the unmanned aircraft and view common analytics during and after the pollination process. The team added weather analysis to the system using a portable ambient weather station to calculate live wind offsets for the unmanned aircraft during pollination.

The system, designed in collaboration with teams 18054 and 18056, controls an unmanned aircraft as it flies through a date palm plantation and pollinates the palms. The computational system mounted on the aircraft uses computer vision and artificial intelligence to observe, model and act on its environment to efficiently and safely complete the pollination process. This includes the identification of the palm trees, the command to release pollen, and the detection and avoidance of obstacles. The system provides the date farmers with a modular product capable of pollinating a field of date palm trees in an industrial-scale farming environment. This allows for significant savings by reducing labor costs and the risk of injury during manual pollination.

The date industry has high costs associated with the palm tree pollination process. Since the palms are only pollinated naturally by the wind, laborers must pollinate the trees in date orchards manually to increase date productivity, a practice that is labor intensive, imprecise and can be dangerous, since it involves reaching heights of 50 feet or greater. Taking inspiration from the successful 2017 senior design project, and in conjunction with teams 18055 and 18056, the team designed a system to standardize and improve pollination practices by using autonomous unmanned aircraft technology and additive manufacturing. The pollen dispensing system allows a set payload of pollen to be deposited on each tree and is mounted directly to a commercially available unmanned aircraft, while retaining all manufacturer flight capabilities. The system also features a quick-detach mechanism to maximize unmanned aircraft up-time and allow for modularity between unmanned aircraft systems. Using additive manufacturing practices like 3D printing, the team has minimized the pollen dispenser’s weight, maximizing pollen carrying capacity and the number of trees pollinated per battery. The multi-team structure of this project has allowed multidisciplinary collaboration in producing a cohesive and modular system.

The OKL4 Hypervisor system is configured by a system-extensible markup language, or XML, file. This XML file defines each virtual machine –its attributes, features and any devices it should have access to. The process of defining a system can be tedious when done by hand. The plugin designed will provide the user with a wizard that allows them to create a hypervisor system step-by-step while also allowing them to save their projects and import previous ones. Major features of the Eclipse plugin include the ability to design a system by picking and choosing options based on valid configurations. This ensures real-time error checking as a user selects options to ensure they are creating a valid hypervisor system.Eclipse allows users of the OKL4 Hypervisor to design their system via a user interface and to produce a bootable image. This eliminates the need for a user to become too familiar with the OKL4 software development kit system XML files that currently define a valid hypervisor system.

Demodulation of high-speed, wide-bandwidth wave forms has typically been performed by ASICs or large CPU clusters. However, development times for these technologies can be long and expensive, especially for ASICs. While general purpose processors are flexible and easier to program, they often lack the throughput required to handle high-speed waveform demodulation. GPUs open up new avenues for flexible demodulators that can be developed quickly and modified and maintained easily. This project seeks to harness the power of GPU processing for high-speed demodulation. To do this, the design uses an industry-leading GPU development environment to optimize the
repetitive processes during demodulation. The system can exploit these processes through parallelization while maintaining a modest implementation loss.

This project tests the performance of the MPSoC device by running software, consisting of an algorithm using cross-correlation, on both the MPSoC’s processing system and on a comparable device to establish a benchmark. The system demonstration uses the same algorithm to perform Wi-Fi signal detection with a peripheral software-defined radio and directional antenna. The resulting signal correlation value and received signal strength indicator, as well as video feed from a camera, are displayed on a monitor to visualize the signal strength at a specific location.The benchmark software uses math functions ported over from a C-language software library device to ensure an accurate performance comparison between the new and current systems. The benchmark software measured the total elapsed time taken to correlate one million input signals with one reference signal. The results from this test were delivered to General Dynamics Mission Systems so it could compare the devices internally to determine if the MPSoC device should be considered for future products.

The growing capabilities of unmanned aircraft have garnered attention from the commercial and defense industries. Demand for unmanned aircraft that can transmit data over long distances in a secure fashion is on the rise. The designed system can establish a secure connection through the user controller and a camera attached to the unmanned aircraft’s flight controller by encrypting and transmitting data over the LTE network in real time. The system can operate as long as both the user and unmanned aircraft are within range of any LTE network, no matter the distance between them. The secure encryption of both real-time video and unmanned aircraft controls limits the threat of third-party attackers learning more about the system’s operations. By connecting the camera to the unmanned aircraft controller, both video stream and commands are encrypted and decrypted over the link to the user’s control station. The system demonstrates advanced, commercially available capabilities for unmanned aircraft to transmit data to a user over unlimited range, contingent on each component being within range of popular LTE networks.