Aptiv is working towards a realistic “Virtual Driving Simulation” to enhance the verification of Advanced Driver Assistance Systems (ADAS) in various scenarios, including corner cases. Students on the Aptiv team will work with existing real-world vehicle data and develop models to replicate sensor (radar and camera) performance in various environmental conditions, providing realistic sensor outputs that drive the behavior of downstream algorithms.

The UofM Campus Farm has a goal of reaching carbon neutrality for all emissions generated and electricity used on-site by 2026. Students on this team will support this goal by developing and implementing an energy management system that allows for a solar-powered produce cooler and EV delivery van to operate as a combined system minimizing or eliminating the need for grid power in the farm’s produce storage and delivery systems.

Every year, Constellation Generation moves thousands of nuclear fuel bundles safely in and out of reactor cores in a challenging underwater environment, which is heavily dependent on humans to ensure reliability. Students on this team will aim to eliminate the human element of this critical task, by adapting a robotic tool for underwater nuclear use, reducing the risk of an error.

Honda envisions a future where unmanned aerial vehicles (drones) work in tandem with cars on the road. Students on this team will build on previous path-planning research, developing and testing various use cases for drones to enhance the automotive driving experience, such as increasing the perception range of automobiles by sensing areas, and streaming data back to the autonomous vehicle.

As the presence of robots in our every day life continues to increase, it will be critical to ensure that they can safely and efficiently interact with humans in an uncontrolled environment. Students on the Honda project will develop algorithms to find feasible, collision-free paths for a Clearpath Jackal robot with a Zed2 camera operating in a crowd of pedestrians.

JPMorgan Chase is the largest bank in the United States, and processes commercial loans in local markets across the country. Students on this team will implement predictive analytics to develop models that forecast important aspects of the business process within the commercial loan division, with a focus on utilizing large language models, and identifying extreme events within a very large and complex data environment.

Koppers is one of the leading producers of railroad ties in North America, which is a product that is stored in large outdoor stacks, that makes inventory time-consuming and difficult. The students will develop an autonomous drone-based camera system to determine the amount of a particular product available within inventory stored in an outdoor, multi-acre site.

The MiTek® Automation Technical Development organization is responsible for creating new and improving existing tools that support factory-based component manufacturing for single and multi-family construction. Students on this team will aim to develop a proof-of-concept automated inspection system to ensure that truss components meet all quality inspection criteria.

Students will develop an AI-driven Virtual Casino Host using OpenAI’s API. The system will track players’ gaming activity, analyze their behavior and preferences, and then tailor and suggest offers to encourage increased activity at the casino.

This project seeks to leap an order of magnitude forward in the domain of software code quality via automated testing. Students on this team will employ Generative AI trained on domain-relevant tests, and continuously educated based on user selection or rejection of generated output, to continuously create system, interface, and unit tests for a complex system.

Historical newspapers are some of the richest, longitudinal sociological data sources available, going back more than 250 years. Accessing content from older sources digitally is challenging. Students on the ProQuest team will use large language models and other ML techniques to develop a proof-of-concept system for reconstructing individual articles from full-page newspaper images.

Students on the Reverie team will design and develop a sound detection and integration feature that accurately identifies snoring and automatically sends a signal to the bed to perform actions to mitigate snoring, cater to individual preferences, and improve sleep comfort for those that have trouble falling asleep.

Subaru wants to implement advanced AI functionality in their Driver Monitoring System (DMS), leveraging passive IR (infrared) research. Students will develop advanced IR sensor applications for driver monitoring and consumer convenience that will be deployed in our prototype WRX, which is actively being displayed throughout the US at various auto shows, racetracks, and extraneous events.

Create an intuitive tool for public health in Michigan, where everyone who needs vaccines gets them! By obtaining real-time data and visualizations, we can support public health in planning and organizing their vaccination efforts.

Photo captions provide context to images, facilitate better communication, and increase accessibility. The student team will use Azure machine learning services, in addition to student developed functions, to build a tool including an API that is capable of creating image captions when given an image input.

Walbridge construction company makes jobsite safety a top priority on all their projects, including their two active jobs on U-M’s North Campus. Students on the Walbridge team will develop a Proximity Warning Alert System (PWAS) that detects people, rather than objects, in the blind spots of mobile construction equipment.

Models used to test control software for appliance development are complex and time-consuming to run. Students on the Whirlpool team will develop machine learning models that replicate first-principles models, reducing run time and accelerating product development.