MISSION:The Michigan Tech Transportation Institute (MTTI) provides the operating structure, resources, recognition, and leadership, in a collaborative environment, that supports research, education, and outreach leading to sustainable solutions for transportation.
Established in 2003, the Michigan Tech Transportation Institute (MTTI) is THE central entity on Michigan Tech’s campus for transportation related research, outreach, workforce development, and educational initiatives that address national and global needs.
This is one of the biggest rail research events in the world, combining the World Congress in Railway Research and International Heavy Haul Association Conference. All Tech faculty with rail research are encouraged to submit.
Research projects awarded to MTTI increased from $1.6 million to $5.5 million over the past year, demonstrating significant growth. To read more about MTTI’s progress and events and activities held in FY24, please see our annual report, submitted to the Vice President for Research. https://www.mtu.edu/research/about/centers-institutes/docs/fy24-mtti-annual-report.pdf
MDOT Director to Visit Campus
The new director of the Michigan Department of Transportation (MDOT), Bradley C. Wieferich, will be on campus for a brief visit on November 4, 2024. MTTI will be leading short tours to various transportation and mobility related facilities for the Director. The visit will be completed with a working luncheon with transportation/mobility research leadership. If any MTTI members/friends have a facility or project they would like to highlight, please contact Pam Hannon.
Newly Awarded Projects
The beginning of new fiscal year has been EXCELLENT for MTTI members with six new projects awarded over the first four months (and more on their way). This includes two new PIs, Vinh Nguyen (MAE) and Shane Mueller (CLS). Congratulations for all PIs with new awards....and welcome Vinh and Shane as new "project leaders".
Standards Development Center for Automated Driving Systems in Inclement Winter Weather, sponsored by the US Department of Commerce/National Institute of Standards and Technology (NIST), PI: Vinh Nguyen
Hot rubber seal coating to survive wet and frozen environments, Minnesota Department of Transportation/National Road Research Alliance, PI: Zhanping You
Project PROSPR: Upcycling Waste Plastic into Thermoplastic Elastomer as a subcontractor to Molten Materials through the National Science Foundation. PI: Zhanping You
Durable rubber-modified epoxy concrete overlay for bridge deck protection and rehabilitation, Michigan Department of Environment, Great Lakes, and Energy (EGLE), PI: Qingli Dai
Soybean-Based Materials for Pavement Maintenance through the Michigan Soybean Council. PI: Zhanping You
Lightweight evaluation, training, and user collaboration for Human-AI Work Systerms in Rail Operations, sponsored by the Federal Railroad Administration. PI: Shane Mueller
Tracks to the Future SYP Completed
Michigan Tech was the lead university on an educational summer youth program Sponsored by the Federal Railroad Administration. The Tracks to the Future (T2F) program ran for three consecutive summers, introducing high school students across the nation to the industry and related opportunities. Each session consisted of two days of virtual learning at the student’s home location, with a travel day to their university location, followed by two days of hands-on activities, tours and field trips to rail sites. Michigan Tech was joined by partners Oregon State University, University of Nebraska – Lincoln, Penn State University – Altoona, University of South Carolina, University of Illinois – Urbana Champaign, Fresno State University, University of District of Columbia, Washington, DC and the University of New Mexico.
Michigan Rail Conference
The 2024 Michigan Rail Conference Pivot to the Future was held August 14–16, 2024 in Livonia, MI with 122 total registrations and over 25 professional industry speakers. The conference, which began in 2013, was established to bring together rail industry stakeholders for meetings and presentations. Michigan Tech is the host university in charge of logistics for the conference.
Project Highlights
Title:Standards Development Center for Automated Driving Systems in Inclement Winter Weather
Sponsor: US Department of Commerce/National Institute of Standards Technology (NIST)
PI: Vinh Nguyen (MAE)
This project aims to conduct the research and demonstration for the establishment of a center to assist standard development organizations (SDOs), manufacturers, and local/state/federal agencies in the performance testing of Automated Driving Systems (ADS) in inclement winter weather. Though ADS testing standards are critical towards promoting quality, public trust, and safety in driving, there is a lack of test methods under inclement winter weather conditions for ADS technology. Through this project, Michigan Technological University (MTU) aims to leverage its automotive, robotics, and measurement expertise to conduct the research and stakeholder engagement necessary to establish a center to inform testing and standards development on ADS operation in inclement winter weather.
To conduct research critical to the establishment of this center, off-vehicle component-level testing of ADS sensors and communications with benchmark calibration targets in inclement winter weather will be conducted. The benchmarking datasets from this center will be supplemented with driving data collected from MTU's existing ADS-equipped vehicle fleet. In addition, automated driving machine learning models for object detection will be used to validate the datasets. The elemental behavioral competency (defined as a single automated driving functionality) of responding to a lane obstruction will be demonstrated to show the effects of inclement winter weather on driving performance. In addition, this center will leverage its research to develop a playbook that will act as a one-stop-shop resource for records of ADS-related test standards. This playbook, resulting educational materials, and ADS in Inclement Winter Weather Awareness Event will be used to provide upskilling resources to the automotive workforce and engage SDOs. Hence, this project will lay the foundation for a specialized center to support standards development and education to advance the robustness of ADS technologies.
Project objectives and execution to lay the foundation for the proposed center.
Center Description
MTU aims to assist SDOs and other stakeholders in the advancement of ADS standards through the creation of an ADS testing center under inclement winter weather. After its establishment at the end of this project, this center will be sustained through MTU’s automotive network and institute-based research infrastructure to conduct the following functions.
Inform SDOs and other stakeholders to advance ADS testing in inclement winter weather.
Provide readily available off-vehicle and driving datasets of ADS perception and communication technology in inclement winter weather.
Facilitate evaluations of machine learning (ML) models to advance ADS robustness.
Maintain an ADS standards playbook for educating stakeholders on ADS test methods.
However, this center will require research and demonstration of ADS performance in inclement winter weather to establish stakeholder interest, which is the purpose of this project. Initially, test methods for evaluating ADS technology under both component-level and ADS-equipped vehicle configurations in inclement winter weather will be conducted. In addition, the center will leverage its research to develop a playbook on ADS test standards to increase public awareness of ADS technology. Therefore, this center will develop test methods for evaluating ADS technology in inclement winter weather to advance US commerce and public interest.
ADS-equipped vehicle at MTU (left) with onboard controller unit (right).
Schematic of responding to lane obstruction.
Title: Lightweight Evaluation, Training, and User Collaboration for Human-AI Work Systems in Rail Operations
Sponsor: US Department of Transportation/ Federal Railroad Administration
PI: Shane Mueller (CLS)
CoPI: Pasi Lautala (CEGE)
CoPI: Elizabeth Veinott (CLS)
This project supports the objective of FRA-HF-003, to develop methods for improving the design, evaluation, and capabilities of work systems involving human users and operators interacting withintelligent software tools. To accomplish this goal, we propose to deliver a series of brief reports providing guidance for lightweight assessments and tools that support human use of intelligent systems (e.g., automation, AI, and ML) drawing on existing research, and to incorporate new human-subjects studies to validate and tailor select approaches to intelligent rail contexts. The project will use methods and approaches previously developed by members of our team for empirically assessing and improving human-AI work systems, enabling us to draw on extensive experience and lessons learned in other contexts. The goal of our approach is to identify lightweight tools that allow rail operators, vendors, and decision makers to be smarter about the AI tools they adopt and use1 throughout all stages of technology planning, development, deployment, and maintenance. We believe these lightweight approaches will be more likely to be adopted by vendors and operators, especially for those without a strong human factors and cognitive systems engineering focus. Our approach will allow FRA to better evaluate how automation and intelligent systems can be used in rail operations to improve performance and safety. The left side of Figure 1 shows seven major needs and goals for deploying intelligent systems in the workforce. The right side shows example approaches or solutions to these needs which will form the basis for our primary research approach.
Figure 1: Seven major needs of organizations deploying intelligent systems.
The seven major needs and example solutions we have identified in Figure 1 cover all phases of technology development, deployment, and maintenance (see Figure 2). We view each to have a critical period for use, insofar as for each type of problem and solution, there are times at which using them may be either too early or too late. Consequently, the best approach depends on the current development-deployment lifespan of an intelligent software system, from early planning to post-deployment. Thus, we propose a complementary research program involving (1) a review of existing approaches for dealing with seven major needs faced by organizations deploying intelligent systems, focused on intelligent rail systems; (2) providing guidance on example solutions to these needs; and (3) human-subjects research base on best practices for assessing human-AI work systems for select approaches to validate and tailor their use to the domains of intelligent rail systems.
Figure 2: Depiction of the lightweight evaluation approaches and their critical periods in the development and adoption of technology.
