The project requires the team to design and test a telecentric optical system with autofocus capabilities for machine vision. The design integrates a liquid lens into an existing telecentric lens system, and incorporates both optical components with a control algorithm. The liquid lens provides the autofocus function without the use of mechanics to adjust the focus. The team had to solve problems such as camera selection, thermal stability, autofocus, maintaining object space telecentricity, and minimizing magnification change with focus. The algorithms accurately report to the user a view of the current focus and provide metrics for the user to decide how much to adjust the focus. The system was tested to meet optical specifications in a mounted environment with the test objective back-illuminated by a sodium source. Images captured of the test objective are processed against the magnification and modulation transfer function requirements and an adjustment to the focus of the system is patched to the liquid lens driver. Automating the focusing of this system allows for more efficient monitoring of any manufacturing process in which machine vision is a tool for qualitative analysis.

The team was asked to review alternative cooling technologies for office buildings and single-family homes in Phoenix, Arizona. The typical structures selected for study were a 50,000-square-foot office building and a 2,500-square-foot single-family residence. Alternative cooling technologies were compared with respect to their energy efficiency, installed cost, and energy cost savings based on a typical utility company electric rate. The baseline cooling system is a standard efficiency direct-expansion cooling unit. The analysis used realistic building models to obtain cooling loads for input into the financial model and a cooling load model for each building type.

The team was asked to design and develop an assisted launch and capture system for unmanned aerial vehicles that can decrease takeoff and landing distance by 10 percent. The team tested the concept by designing and building a subscale prototype that uses compressed air to provide extra thrust sufficient to decrease the launch distance. For capture, the system slides the launch box from the center of the runway and uses a cable to catch the plane as it touches down. The cable provides the resistance to slow the plane down and stop it within the decreased landing limit. The modular design allows the system to launch and capture the plane efficiently without damaging it. The design can be scaled to accommodate full-size unmanned aerial vehicles and used for future testing.

Universal Avionics sometimes receives incorrect or inconsistent data from external inertial measurement and magnetic sensing units installed in its own attitude and heading reference system. The externally sourced units consist of gyros, accelerometers, and magnetometers and are classified as microelectromechanical system sensors. The sponsor has no way to test the external systems in-house and has to return them to the supplier. The team’s objective is to create a test system that enables the sponsor to validate the functionality of the external units in-house. The system design includes communication with the units via a software interface, electrical power and communication, and a mechanical test system that simulates the environment of the unit installed on an aircraft. The environmental test system can be calibrated in the pitch, roll, and yaw directions and is strapped to a cart for mobility. The roll and pitch are adjusted using a precision tilt and turntable with the sensors mounted on top of the structure. The desired yaw direction is achieved by aligning pins on the cart with the sponsor’s sight compass. Communication with the sensors is regulated via a graphical user interface and based on a custom circuit board containing three microprocessors and power regulators.

The purpose of this project was to develop portable remote sensor module to aid in troubleshooting by sensing real-world signals and recording the information over a selectable time period. The sensor module detects and records electrical current, temperature, light interruption, motion, and the presence of fluids. Recorded information is downloaded to a portable device such as a laptop or notebook, where users can view the data on an easy-to-understand graphical user interface. The sensor module uses the Texas Instruments MSP 430 microcontroller and several more of the company’s precision operational and instrumentation amplifiers. The team designed the module to be marketable, with 15 percent profit margin as a minimum goal.

The project required the team to create a computational way to analyze efficiently the data generated by the build process of Honeywell’s direct metal laser sintering additive manufacturing unit, which uses an EOSINT M80 3-D printer. The temperature of the material being sintered, and the rate at which the temperature changes, must be verified during printer operation. The high-powered laser used in the sintering process causes a major change in the temperature of each layer of the build. Printer data input to the software enables users to characterize the build in terms of heating and cooling rates, and peak temperature of material. The software notifies the user that analysis is complete and provides numerical and graphical data output for review and storage.

Residue on kitchenware in a dishwasher influences the decision to continue using a brand of detergent. The objective of this project is to design a system to quantify this residue and provide a cleanliness grade that mirrors how a human would visually grade the kitchenware. The team focused on measuring spotting and film left on knives, because knives can be modeled as mirrors. The team measured and graded cleanliness using distinctness of image, a metric that uses contrast degeneration to determine image degradation. For example, perfect mirrors reflect black and white stripes in sharp contrast, but a surface with degraded reflective properties blurs the distinction between the same black and white stripes. The team projected a striped pattern onto the knives and measured the degradation of distinction between segments imaged, investigating the correlation between cleanliness and distinctness of image.

Caterpillar tests electrical systems by simulating system failure using breakout boxes that communicate with various equipment subsystems via an electronic control module. The sponsor charged the team with designing an automated breakout box with a laptop-based graphical user interface to report results to the user. The designed system consists of an interface box in the cab, and a main box by the electronic control module being tested. The laptop connects to the interface box, where system power is provided by equipment being tested, and houses a microcontroller, LED indicators, and a memory card. The main box houses 18 printed circuit boards, each consisting of 14 relay boards, a power board, a signal control unit, and a fault rail. Relay boards control signals from the electronic control module, sending them through by default or sending signals to the fault rail, which performs tests and records results. The signal control unit facilitates testing and the power board distributes power from the truck to the test system. The entire automated breakout system is controlled by a custom user interface that allows tests to be performed at the click of a button. Tests can also be put in a queue and run without an operator.

Thermomechanical fatigue analysis studies how cyclic thermal loading and large operating temperature gradients cause material fatigue. Test equipment often uses induction as a heat source, which makes systems expensive and difficult to analyze, so the team set out to find an alternative heat source. Using combustion heating and forced convection cooling, the team has devised a way to simulate real, accurate thermomechanical fatigue conditions at a fraction of the cost. The design optimizes the geometry of a nickel-based superalloy specimen to achieve a target stress of 80,000 psi and temperature of 2,100 degrees Fahrenheit when heated and cooled simultaneously. The method involves heating the top face of the specimen with a high-temperature oxy-propane torch while a vacuum draws air through a slot in the specimen at extremely high velocity. Simultaneous heating and cooling produces a temperature gradient between the top and bottom of the specimen, causing it to expand and compress at the same time, thus creating the desired compressive stress. This method will allow the sponsor to efficiently and cost-effectively explore thermomechanical fatigue properties of proprietary materials for future use in jet turbine engines.

Aircraft bleed air at high temperature and pressure comes directly from the engine before jet fuel is added. This air is used to perform aircraft functions such as pressurizing the cabin and running the air-conditioning unit. Contaminants in bleed air can damage aircraft components, and the goal of this project is to create a system to detect contaminants when they reach significant levels. The mechanical subsystems were designed to reduce the extreme conditions of the bleed air stream to within the functional range of the sensor used to detect the contaminants. The electrical subsystem communicates information about contaminant levels to a microcontroller for processing. The microcontroller also provides an indicator to the user interface for aircraft operators.