Electrical and Computer Engineering Commons^™

A Near-Field Communication (Nfc) Multi-Sensor Node With Optimized Read Range And Adaptive Power Management For Remote Monitoring, Rishin Patra, Hilary Scott Nkimbeng Cho, Jin W. Choi

Michigan Tech Publications

This paper presents the design of a batteryless near-field communication (NFC) multi-sensor node with an integrated adaptive power-management system for sensing applications. The work focuses on harvesting energy from a 13.56 MHz NFC field to power an ultra-low power sensing platform. The design consists of the TI RF430FRL152H, an integrated NFC transponder with an embedded MSP430 microcontroller core and ferroelectric random-access memory (FRAM) non-volatile memory. The system combines an ISO/IEC 15693 NFC front end, a tuned loop antenna for optimized power harvesting, and multiple analog and digital sensor interfaces, and a firmware architecture for intermittent harvested energy operation. The aforementioned …

Dynamic Deep Prompt Optimization For Defending Against Jailbreak Attacks On Llms, Doniyorkhon Obidov, Honggang Yu, Xiaolong Guo, Kaichen Yang

Michigan Tech Publications

Large Language Models (LLMs) demonstrate impressive capabilities across many applications but remain vulnerable to jailbreak attacks, which elicit harmful or unintended content. While model fine-tuning is an option for safety alignment, it is costly and prone to catastrophic forgetting. Prompt optimization has emerged as a promising alternative, yet existing prompt-based defenses typically rely on static modifications (e.g., fixed prefixes or suffixes) that cannot adapt to diverse and evolving attacks.

We propose Dynamic Deep Prompt Optimization (DDPO), the first jailbreak defense based on deep prompt optimization. DDPO uses the target LLM’s own intermediate layers as feature extractors to dynamically generate defensive …

Remote Sensing Of Dynamic Ground Motion Via A Moiré-Based Apparatus, Adrian Ali Moazzam, Nontawat Srisapan, Gregory Waite, Durdu Guney, Roohollah Askari

Michigan Tech Publications

Highlights: What are the main findings? A Moiré-based optical apparatus enables long-range, non-contact ground displacement measurement in hazardous environments. Controlled indoor and outdoor experiments demonstrate reliable detection of dynamic and seismic-like ground motions with sub-millimeter resolution. What are the implications of the main findings? System performance is evaluated under atmospheric turbulence and wind, revealing key limits and mitigation strategies for field deployment. The proposed approach provides a low-cost, scalable complement to traditional seismic and geodetic monitoring techniques. Ground-based remote sensing of seismic and geophysical displacements remains a major challenge due to environmental hazards, signal attenuation, and practical deployment limitations of …

Ainet: Integrating Mamba And Cbam For Enhanced Camouflage Object Detection, Henry O. Velesaca, P. Andrea Mero, Abel A. Reyes-Angulo, Angel D. Sappa

Michigan Tech Publications

This paper introduces AINet, a novel deep learning architecture designed for detecting camouflaged objects in complex and diverse environments. The objective of this work is to design an end-to-end camouflaged object detection architecture that simultaneously captures long-range dependencies and refines subtle camouflage cues, improving segmentation accuracy and boundary delineation across both standard COD benchmarks and real-world agricultural scenarios. AINet leverages the strengths of Mamba, an efficient sequential state model for capturing long-range dependencies, and the Convolutional Block Attention Module (CBAM) for feature refinement through attention mechanisms. Detecting camouflaged objects is a significant challenge across a wide range of real-world applications, …

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 12, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 16, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 20, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 34, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 76, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 11, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 13, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 14, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 15, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 17, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 18, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 23, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 26, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 28, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 30, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 35, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 36, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 37, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 77, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 77 Lidar, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 22, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Wads 3d Object Detection Dataset (Wads-3d) - Sequence 24, Yiming Yang, Jeremy P. Bos

WADS-3D

WADS-3D is a standard KITTI-format dataset designed for analyzing 3D object detection robustness in severe winter weather. Addressing the need for granular analysis of perception failure in snow, the dataset includes 4,100 frames and over 30,000 labeled car instances captured during heavy snowfall and winter-storm conditions. Derived from the large-scale multimodal WADS dataset, WADS-3D enables researchers to move beyond qualitative weather assessments and systematically evaluate how varying heavy snowfall intensities impact a wide range of neural network detectors.

Silent Sabotage: Internal State Triggered Backdoor Attacks On Llm-Powered Robotic Systems, Doniyorkhon Obidov, Shivayogi Akki, Tan Chen, Kaichen Yang

Michigan Tech Publications

The integration of Large Language Models (LLMs) into robotic control systems is enabling a new generation of autonomous agents capable of complex reasoning and planning. While this paradigm shift accelerates progress, it also introduces novel security risks that remain largely unexplored. Current research into LLM backdoors has focused on attacks triggered by external stimuli, such as specific words, visual objects, or environmental states. These attacks, while potent, overlook a more insidious class of vulnerability where the trigger is internal to the agent’s own operational logic. This paper presents the first comprehensive study of history-based backdoor attacks on LLM-powered robotic systems. …

Multi-Level Energy Optimization For Connected And Automated Vehicles: From Cooperative Multi-Vehicle Control To Individual Powertrain Management, Pruthwiraj Santhosh

Dissertations, Master's Theses and Master's Reports

The transportation sector currently accounts for nearly 30% of global energy consumption, necessitating urgent advancements in vehicle efficiency to meet Net Zero targets. Leveraging connectivity and automation, this dissertation proposes and validates methodologies to reduce the energy consumption of light-duty vehicles at both fleet and individual levels.

First, a validation framework is developed to bridge the “simulation-to-real world” gap in Cooperative Automated Vehicle (CAV) research. Moving beyond virtual simulations, the study establishes a methodology for physically validating centralized control architectures via a custom Cellular V2X network. By synchronizing vehicle-powertrain models with physical test vehicles, the framework successfully orchestrates complex arterial …

Design, Modeling, And Experimental Development Of Nanoscale Confinement Structures On Planar Silicon-Based Microelectrode Arrays For Single-Entity Electrochemical Sensing, Parinaz Eskandari

Dissertations, Master's Theses and Master's Reports

Electrochemical sensing is widely used for chemical and biological detection due to its high sensitivity, label-free operation, and compatibility with miniaturized electronic systems. However, conventional microelectrode platforms operate in an ensemble-averaged regime in which the measured current represents the collective response of many molecules interacting with the electrode surface. This ensemble averaging masks localized nanoscale electrochemical events and limits the ability to detect rare interactions, such as single molecules or nanoparticles. Achieving single-entity electrochemical detection therefore requires strategies that confine electrochemical reactions to nanoscale regions while maintaining compatibility with scalable planar microfabrication.

This dissertation investigates nanoscale electrochemical confinement on planar …

Experimental Rate Feedback Control Of A Model-Scale Hourglass-Shaped Heaving Point Absorber, James R. Halverson

Dissertations, Master's Theses and Master's Reports

Buoy geometry greatly affects a point absorber wave energy converter's dynamic response to waves. Finding the optimal buoy shape and control method remains an open research area focused on maximizing the conversion of wave kinetic energy into electricity. This work presents an experimental comparison of closed-loop energy extraction between a cylindrical and a truncated cone buoy, both with the same submerged volume, across various wave frequencies and amplitudes. To ensure a fair comparison, the optimal rate feedback gain is calculated for each buoy at each wave condition. Multiple metrics, including power output, capture width, and actuator force, are used to …