������Ƶ

Dr. Diane Peters is an Associate Professor of Mechanical Engineering. She has a long history of mentoring undergraduate students both while working in industry and later when she went into teaching engineering. She is the faculty advisor for many student organizations on campus. She has published papers with undergraduate student authors on many different topics. She has a strong interest in control systems and their applications, including how they can be used to develop autonomous vehicles.

Dr. Rui Zhu, the co-P.I. for this proposal, is an Assistant Professor of Computer Science. His research aims at the security, privacy, and performance in wireless network systems and applied machine learning algorithms. His research interests span a broad range of problems, including PHY-layer security in wireless networks, millimeter-wave communication security, optimization of machine learning models, autonomous mobile robots (AMRs), and vehicular networks. His experience in guiding and facilitating undergraduate research in REU programs in the past greatly contributed to the success and growth of the students.

Dr. Daniel Ludwigsen is an Associate Professor of Physics and Department Head for the Department of Natural Sciences. He has mentored many undergraduate students who have conducted undergraduate research theses and other independent projects, some of which have resulted in publications. Dr. Ludwigsen grew up in rural Illinois in an area where few students went on to college; as such, he understands the students who come from rural backgrounds and can serve as an effective mentor to them.

Dr. Michael Farmer is a Professor of Computer Science and Department Head. Dr. Farmer has a strong commitment to undergraduate education, which led him to join the faculty at ������Ƶ. He has published several papers with undergraduate student authors. Dr. Farmer was a first-generation college student, and has a unique understanding of the challenges faced by students who come from similar backgrounds.

Dr. Jungme Park is an Associate Professor of Electrical and Computer Engineering. Dr. Park’s specialties are Computer Vision and Artificial Intelligence (AI). She conducted many AI-related projects while working in academia and the automotive industry. At�� ������Ƶ, her research interests lie in environmental perception for Autonomous Driving (AD), driver awareness detection, Sensor Fusion, and AI application development. Based on her research work experience, she strongly believes in the synergy through collaboration in research.

Dr. Mehrdad Zadeh is a Professor of Electrical and Computer Engineering. Dr. Zadeh is an advisor for multiple student teams in the area of autonomy. He is interested largely in the areas of Cyber Physical systems, including VR/AR and Automated Driving. Dr. Zadeh has advised multiple students at both the graduate and undergraduate levels in research on autonomous driving, and has developed several new courses on relevant areas.

Dr. Theresa Atkinson is a Professor of Mechanical Engineering and the director of Kettering’s Crash Safety Center. Prior to receiving her Ph.D., she worked for General Motors in design and testing of thermal systems. After receiving her Ph.D., she established an engineering consulting business and a nonprofit that provided support for crash survivors. Her interactions with students outside of the classroom include teaching in the Lives Improved Through Engineering and Science (LITES) summer camp each year and mentoring senior mechanical engineering students, masters students, and medical students in research. Dr. Atkinson’s research area can be broadly described as injury prevention, with focus areas of vehicle safety and orthopaedic trauma research.

Dr. Javad Baqersad is an Associate Professor of Mechanical Engineering at ������Ƶ. He has been working with undergraduate students as part of capstone projects as well as research theses. Some of these capstone projects include designing and building autonomous RC cars; others involve design and construction of a drone for structural health monitoring. Yet others include the KU-iTire project and control of highly flexible rotating structures.��

Dr. Chinwe Tait is an Assistant Professor of Electrical Engineering in the Department of Electrical and Computer Engineering (ECE). Dr. Tait spent five years working on cutting-edge technology as a Sensor Engineer in the automotive industry, and brings that industry perspective to her research and mentoring of students. Dr. Tait is the first in her family to obtain a graduate-level degree, and has mentored numerous underrepresented minority (URM) students ranging from high school age to university level.

Dr. Lisa Gandy is an Associate Professor in the Department of Computer Science at ������Ƶ. Prior to joining ������Ƶ, Dr. Gandy was an assistant and then associate professor at Central Michigan University and served as chair of the department. Dr. Gandy’s research interests focus on natural language processing and text mining. Dr. Gandy has mentored the Women in Technology student organization at CMU for 10 years and is currently mentoring the student chapter of ACM at ������Ƶ.

We will be adding and updating projects each year as technology advances! The initial projects that students may have the opportunity to work on are:

Comparison of Control Algorithms on a Small-Scale Vehicle Testbed (Mentor: Dr. D. Peters)

Many different approaches have been used to control autonomous vehicles. Which approach is best depends on what the criteria are, what type of sensor data is available, and on the operating domain of the vehicle. In this project, students will utilize a small-scale vehicle testbed to compare different algorithms in a variety of different environments. The vehicles are based on RC car parts, with a custom chassis to mount sensors and controllers, and are 1:16 scale. The research questions in this project will focus on the tuning of algorithms, comparison of different algorithms, and for more advanced students, the development of new algorithms. In regard to tuning algorithms, students will determine if a procedure based on a solid theoretical basis can be developed and how it compares to empirical methods in existence.��

Enhanced Interior and Pass-By Noise Analysis of Vehicles (Mentors: Dr. D. Peters, Dr. J. Baqersad, Dr. D. Ludwigsen)

Noise has been used as an essential means of sensing and communication in vehicles, particularly for those with limited vision. In addition, with new developments in electric and autonomous vehicles, the noise signatures of vehicles will change due to the differing powertrain as well as the placement of sensors. Understanding the noise characteristics of these vehicles is critically important in mitigating the negative impacts of this noise as well as in addressing the ability of those with limited vision to detect the presence of vehicles. This project will measure the noise characteristics of an electric vehicle using advanced equipment, including a laser vibrometer and Head Acoustics tools, on the university’s test track. Various sensor configurations will be tested at different speeds, and the resulting data will be analyzed to identify noise characteristics based on volume and frequency. This research aims to provide valuable insights into optimizing EV noise profiles, enhancing both safety and comfort, and supporting inclusivity for visually impaired individuals.

AI-based Misbehavior Detection in V2X Wireless Networks (Mentor: Dr. R. Zhu)

Vehicle-to-Everything (V2X) technology enables real-time data exchange between vehicles and infrastructure, enhancing safety and situational awareness. However, the dynamic and data-intensive nature of V2X systems introduces significant security challenges, particularly for resource-constrained devices like On-Board Units (OBUs). Traditional cryptographic methods struggle to meet the real-time demands of detecting and mitigating threats, underscoring the need for advanced, data-driven security solutions. This project aims to develop AI-based misbehavior detection techniques tailored to V2X networks. Students will collect real-world data from Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) interactions using V2X simulation tools and cutting-edge equipment like software-defined radios and Cohda On Board Units. The data will be used to train machine learning models capable of distinguishing legitimate from malicious traffic, ultimately building a comprehensive database of V2X network attacks and enhancing the security of autonomous vehicle systems.

Determining Key Factors in Vehicle Safety/Crash Avoidance for Autonomous Vehicles (Mentor: Dr. T. Atkinson)

As autonomous vehicles (AVs) become more prevalent, the nature of vehicle crashes and injury risks will evolve. Changes such as non-forward-facing seating in AVs pose new safety challenges, as current restraint systems are optimized for traditional seating. Additionally, crashes between heavier battery-powered vehicles and older, less-advanced vehicles may increase injury risks for occupants of the latter.�� This project aims to identify emerging safety concerns by analyzing national crash data and conducting crash experiments at ������Ƶ's Crash Safety Center. Students will learn research methods, including literature review, study design, and data analysis. They will explore topics like vehicle structure and occupant injury, collaborating with graduate students and participating in crash tests.�� This research will provide insights into injury prevention and safety system improvements as AVs integrate into the current vehicle fleet.

Intelligent Tire (KU-iTire) for Improved Ride and Handling of Autonomous Vehicles (Mentors: Dr. J. Baqersad, Dr. M. Zadeh)

With the advancement in autonomous vehicles and driver-assisted systems, there is an increased need for intelligent tire systems that can accurately assess vehicle and road conditions in real time. Current vehicle control systems estimate the tire-road contact parameters indirectly, making them vulnerable to sudden road changes. This project proposes KU-iTire, an intelligent tire model that uses physical-informed machine learning to predict braking and traction requirements based on strain and acceleration data from sensors embedded in the tire. A finite element model is used to replicate the test data virtually and is implemented to study the non-linear behavior executed by a rolling tire. This data will be used in a physics-based machine learning algorithm to predict real-time contact patch situations to the vehicle GPU in order to generate the optimum traction or braking force required for the vehicle’s safe operation. This research integrates tire physics with machine learning to improve safety and performance by optimizing traction and braking forces, providing early warnings of hazardous road conditions.

Handling Uncertainty in Complex Environments (Mentor: Dr. M. Farmer)

Autonomous vehicles, as with manned vehicles, must manage uncertainty from ambiguous sensor measurements, unclear road conditions, and uncertain sensor availability, etc. While traditional methods like Bayesian and Dempster-Shafer reasoning address uncertainty, human cognitive models offer a unique advantage: the ability to "forget" irrelevant information. This is particularly useful in dynamic scenarios, such as adapting to long-term road closures, where outdated data may hinder performance. This project explores how human cognitive models can enhance uncertainty management in autonomous vehicles. By integrating insights from psychology and artificial intelligence, students will design and test simulations that incorporate forgetting mechanisms. These simulations will compare the effectiveness of cognitive models against traditional methods in handling evolving conditions. The research aims to improve decision-making in uncertain environments, providing a novel approach to enhancing the reliability and adaptability of autonomous systems.

Robust environmental perception system for Autonomous Driving using visual sensors (Mentor: Dr. J. Park)

In Autonomous Driving (AD) vehicles, in-vehicle sensors perceive the surroundings of the host vehicle. Environmental perception is considered one of the most important modules in AD, enabling vehicles to detect obstacles, predict their trajectories, and communicate with onboard controllers. While camera sensors, aided by Deep Neural Networks (DNNs), excel in object detection and classification, they face challenges in low-light conditions and lack precise range information. To address these limitations, advanced sensors like stereo and thermal cameras are being explored to enhance perception capabilities. In this project, students will gain hands-on experience in developing and evaluating computer vision systems for autonomous vehicles. They will calibrate various cameras, collect and process visual data, and implement object detection systems using different sensor types. By comparing these systems' performance, students will deepen their understanding of environmental perception and its role in autonomous driving, contributing valuable insights into the effectiveness of different visual sensors.

Obstacle Avoidance using Monocular Depth Estimation (Mentor: Dr. M. Zadeh)

Depth estimation is an integral part of autonomous driving for collision prevention, but traditional methods relying on expensive sensors like Lidar or stereo cameras limit widespread adoption.�� This project explores a more cost-effective approach using 2D image data for real-time depth estimation.�� Instead of generating full-scene depth maps, which are computationally intensive, the proposed method focuses on detecting obstacles and estimating their depth directly. Students will develop an end-to-end deep learning model that identifies obstacle boundaries and maps them to depth information using the KITTI dataset.�� By streamlining the process, this method aims to be faster and more practical for real-time applications, particularly in shared mobility systems where quick decision-making is essential.�� This research will contribute to advancing efficient, affordable depth estimation techniques, improving safety and accessibility in autonomous vehicles.

Building Trust in Autonomous Vehicles using Object Explanation with Diverse Voices (Mentor: Dr. L. Gandy)

A major hurdle regarding autonomous vehicles is for the user to have trust in the autonomous system’s decisions and actions. Prior research demonstrates that user trust is strengthened when the system provides an explanation for its actions. With the advent of powerful vision recognition systems and large language models, it is now possible for the environment that the vehicle is reacting to to be described to the user in real time. In this project, students will develop two user interfaces for an AV simulator: one providing low-level explanations and another offering high-level explanations using the BLIP-2 system, which integrates vision and language processing. A study will measure user trust across three scenarios: no explanation, low explanation, and high explanation. Additionally, the project will explore the impact of explanation format by comparing text, male, and female-generated speech. This research will advance understanding of user interaction and trust-building in autonomous systems.

Investigation of Metal-oxide Nanosensors for Early Detection of Electric Vehicle Battery Failure (Mentor: Dr. C. Tait)

As vehicle electrification is becoming more prevalent, safety in regards to Li-ion batteries is becoming more of a concern. When a defect develops in the battery cell (or outside of the battery itself), flammable gas generation, extensive heating, and, eventually, a fire and explosion may arise. Recent research points to gas sensors as a possible early detector of battery failure by sensing gas emissions developed during initial leakage. In this project, the student will explore pre-fabricated nanosensors through a structured four-step process. They will use advanced equipment like the Environmental Scanning Electron Microscope (ESEM) and X-ray Photoelectron Spectroscopy (XPS) to confirm sensor composition, measure sensor responses to gas emissions, and convert this data into gas concentration metrics. The findings will help improve early detection methods for battery failure, enhancing the safety of electric vehicle batteries.