48440 research outputs found
Sort by
TORI: Tour-Optimized Robotic Intelligence
No full textTori is an intelligent robotic agent designed to be an autonomous tour guide robot of WPI’s Unity Hall. Using its multi-modal sensing and advanced reasoning capabilities, Tori enhances WPI visitor engagement by guiding users to their desired destinations in Unity Hall on request. The navigation system dynamically adjusts the robot’s path to avoid obstacles and plan safe routes. A voice command system and touchscreen GUI allow for seamless and engaging human-robot interaction methods. In addition, an on-board Large Language Model (LLM) allows users to ask Tori for information about WPI’s campus or programs and receive accurate and informative responses. Tori traverses all five floors of Unity Hall, complete with custom-trained vision pipelines to allow her to use the elevators
Enriching the Education Stoplight
No full textThis project collaborated with Fundación Paraguaya’s Education Stoplight program to provide Cerrito Agricultural School students with tools to improve their own education. Using teacher interviews, student surveys, and classroom observations, effective methods of learning were identified. Based on these results, the “Solutions Bank” website was enriched through adding interactive activities designed to inspire student action. By teaching classes and sharing written instructions, understanding of self-actualized improvement was increased, and an outline was made to continue growth after the project’s completion
Supporting Boricuá’s Network with Improved Data Systems
No full textFood sovereignty is emerging as a priority agenda in Puerto Rico due to climate change, economic instability, and supply chain disruption and communities are therefore striving for greater control of their food systems. The challenge for Puerto Rican farmers is the lack of accessible, well-organized data. This project addressed this issue with the Organización Boricuá de Agricultura Ecológica farm network by developing a farmer membership system that integrates survey data, maps farming activities, and facilitates communication among stakeholders. By building a database using Google applications, including Google Forms, Sheets, My Maps, Sites, and Drive, we will support Organización Boricuá de Agricultura Ecológica to have a more solid network of farmers and enhanced coordination
Benefits and Barriers to Right of Way Rain Gardens
No full textThe US EPA estimates ten trillion gallons of untreated stormwater enter US water bodies every year (Denchak, 2022). Our goal was to help central Massachusetts municipalities protect water sources with the natural water capture and filtration of rain gardens. Realizing our goal took nine expert interviews, three site visits, and coordination with our sponsor, Emma Williamson, Auburn Town Engineer, and the Central Massachusetts Regional Stormwater Coalition. Through our research we developed a guide on building rain gardens, accompanying educational materials, and built a rain garden at Auburn’s Senior Center. Our garden is Auburn’s test case, with potential success helping persuade Auburn government and residents about the utility and value of rain gardens
Transcranial magnetic stimulation at the microscale
No full textNoninvasive neural stimulation techniques allow researchers and medical professionals to modulate brain activity without the use of implants or other invasive procedures. Transcranial Magnetic Stimulation (TMS) is an example of such a technique, which entrains neuron activity through the use of an external magnetic coil. This project will compare cortical and rat ionic channels in a conductance based model to measure neural entrainment
Investigation of the Dynamic Beam Equation
No full textIn this MQP, the Euler-Bernoulli beam equation is solved by reducing the single 4th order PDE to a system of two 2nd order PDEs and then investigating a similarity transformation applied to the system. The Magnus expansion is initially used to approximate the infinite series solution to the ODE system that results from the similarity transformation. This analysis indicated that an exact power series solution could be found governing the motion of the beam for a sudden disturbance. Exact formulas for the power series were found and shown to be equal to the power series for Fresnel and integrals. Visualizations for both the power series approximation and the fundamental solution, the Fresnel integrals, are given. Finally, an alternative classical derivation is given using the Fourier transform and Laplace transform
Characterization and Post-Processing of Medical-Grade Quantum Dots
No full textQuantum dots, nanoparticles which have quantum mechanical properties, have many applications within industries such as optics and energy. Valuable traits such as discretion of energy levels allow them excellent fluorescence in imaging, leading to investigation within biological applications. Recent developments in hydrophilic quantum dots allow for use within these systems, but inconsistencies in downstream processing hinder their output. Our team designed three methods to post-process quantum dots through thermal annealing. Further characterization of absorbance was studied using UV-Vis spectrometry compared to peak emission spectra provided by our sponsor to measure the effects of batch versus continuous annealing. Batch processing showed trends which reflected previous literature, but continuous methods showed positive trends that would be best suited for future work
Advanced Design of an Impactor for Testing and Evaluating Novel Energy-Absorbing Composites for Transportation and Defense Applications
No full textThis project presents an enhanced design of a high-velocity pneumatic impactor for testing and evaluating novel energy-absorbing composite materials. The new design provided capabilities for easy and automatic use, while also improving accuracy and reliability of data collection. This was accomplished by using an intricate solenoid trigger, a remote start, a robust platform for specimen placement, and interchangeable impactor heads to accommodate various sample geometries. The mechanical components of the impacting system were integrated with a Raspberry Pi Pico, Micro-Python, a PCB board, and a Vishay signal conditioning amplifier connected to an oscilloscope. In addition, a high-speed camera was implemented along with Digital Image Correlation software to collect and process data. These aspects combine to allow novel research to be conducted into the mechanical properties of energy-absorbing materials
Developing an Autonomous Drone for Avalanche Search and Rescue
No full textWhile state-of-the-art technology has been developed to aid in avalanche search and rescue, there is still a need to reduce search time in order to increase the amount of live rescues. Our MQP team’s solution to reducing search time is to use a drone which can fly over the avalanche debris field and locate where the person is buried with a 3m accuracy. By characterizing how avalanche beacons behave with noise, designing a proper mount for the sensor, designing a robust system architecture to transfer the data collected, and implementing search algorithms, the design was tested for consistency and reliability
Fabricating Fruit Textures by Continuous Flow Spherification
No full textTexture can be a critical factor in a product’s success on the market, providing an additional layer of consideration beyond taste. Despite this, it is difficult to describe or quantify what exactly inspires positive feelings from a consumer regarding texture, and thus difficult to innovate in the field. Foodberry, a food science company, takes inspiration from real fruits to create healthy snacks with a satisfying texture, one that imitates the encapsulated liquid and gel-like skin of berries. The application of this concept, brought about through sodium-alginate based “spherification” technology, was further explored in this report. A hierarchical design that emulated a raspberry made up of many smaller droplets was proposed and created through a continuous spherification set-up that utilized droplet formation technology. Various iterations of the set-up were explored, including droplet formation strategies with in-line spherification, an additional oil medium that separated droplets, and the use of air to form alginate droplets that were later crosslinked in a separate calcium bath. Once a method was decided upon, various parameters were adjusted, including the concentration of calcium and alginate in solution, the flow rate of the system, and the residence time allowed for the spherification process. Using agricultural ripeness standards, the fabricated droplets and berries were then evaluated texturally. The average diameter of the berries was tested, as well as the strength of the berries when both pierced and crushed. A set-up in which air was utilized to form alginate droplets separately from calcium was found to be most effective. The characterization data showed that a higher percentage of alginate in solution would create stronger berries that more closely resembled the real fruits. The formation of air bubbles in the droplets could be minimized by manipulating the flow rate. Regarding residence time, the small size of the droplets meant that full crosslinking occurred quickly, and after a short time, increasing time spent in calcium solution yielded no further benefits. This proposed product may be utilized by Foodberry to expand their product range, providing a different textural experience for the consumer than most pre-packaged snacks