Torque Signature is a machine learning solution consisting of a model, using the Optuna software library, for predicting the valve position based on streamed or static torque data. The predicted position is validated for accuracy against prescribed metrics. Research Area: AI, ML, NLP
The project provided a mechanism for verifying and validating 3D models, integrated within the BRL-CAD 3D modeling program. Previously 3D modelers had to devise their own list of commands to verify and validate their models. Research area: Graphics, visualization and computational fabrication.
A web application that can communicate to any sensor/microservice via a TCP/IP connection. Displays the sensors/microservices information on the frontend for users to see and manipulate, dynamically refreshing to output new information. Research Area: Human-computer interaction.
The project created a system, using machine learning and natural language processing, to detect unconscious bias within syllabi that can potentially discourage underrepresented minorities. Research Area: AI, ML, NLP.
The application, Sandia Power Analysis Tool (SPAT), allows for a flexible, modular, and efficient approach to analyzing output from a power analyzer, to uncover inconsistency. SPAT was developed using LabVIEW VI file and python scripts. Research area: Data science.
The project involved building a machine learning solution to effectively classify any item from a supplier into the Varis Ontology, resulting in a single and solid classification system. Research area: AI, ML, NLP
Manual lifting at the workplace is a common cause of worker injury. The team created a system that analyzes the worker’s lift form, and educates them on correct lifting techniques, mitigating injury in the workplace. Research area: Computer vision.