Automation and robotics in the workforce present significant ethical challenges for the future. This research examined how "technological disemployment" may go beyond meaning precarious labor toward outright obsolescence of human labor. There is an emphasize on the socio-economic consequences of automation and raising concerns about justice and the future of human  work. There is also a focus on the micro-level of robotic decision-making, alongside underscoring the importance of human value rather than purely efficiency-driven goals. Industrial integration studies highlight the practical transformation of production systems through AI and robotics, noting gains in productivity and safety, but risks such as job displacement, cybersecurity threats, and organizational responsibility gaps. Collectively these findings argue that ethically integrating automation requires not only anticipating labor market disruptions but also embedding moral accountability for the companies integrating these automatic systems and to ensure human well-being remains central in increasingly automated workplace.

The findings emphasize the importance of ethical integration of automation, how it impacts economics and human labor; additionally detailing on the framework and design of robots, showing their interaction and affect on human work. 

Visible Light Communication (VLC) is a promising candidate for next-generation indoor wireless networks due to its high bandwidth potential and immunity to traditional radio frequency interference. However, VLC has a fundamental limitation; it requires line of sight.  Visible light signals cannot pass through walls or obstacles, meaning blockages significantly degrade connectivity and reliability in realistic indoor environments.

Reconfigurable Intelligent Surfaces (RIS) offer a potential solution. By acting as mirrors, RIS can redirect optical signals around obstacles, mitigating LoS limitations and improving coverage. While the general idea suggests significant performance gains, system behavior under practical conditions remains mostly unexplored.

We previously worked with Monte Carlo simulations in MATLAB to model indoor VLC environments with varying obstacle configurations, transmitter fields of view, and RIS placements. These simulations provided insight into how system parameters influence outage and link reliability. However, simulation results require real-world validation.

The current phase of this research transitions from theory to experimentation. A scalable testbed has been developed using Raspberry Pi nodes configured as transmitters and receivers, coordinated through a centralized controller. Although WiFi is currently used as a proxy communication medium, this framework establishes the methodology for collecting dynamic performance data in realistic setting with a variety of conditions and set ups. By systematically adjusting node placement and measuring throughput and collision behavior, the testbed generates data to validate and refine simulation models.

In modern communications systems, the prevalence of pervasive wireless communications creates a growing need for privacy, hence motivating covert transmission where messages are exchanged without detection by an adversarial observer. Prior work has demonstrated that a positive rate of covert communication is achievable through the presence of an uninformed jammer; however, the channel conditions required to enable such transmission are often described in abstract and unintuitive terms. This work proposes a simplified and easily identifiable characterization of the class of distributions for the random jamming that permit positive-rate covert communication. Analytical derivations and simulation-based experiments are conducted on commonly used distributions in wireless communication to examine their defining statistical properties and their influence on covert performance. The results reveal a strong relationship between the shape of a channel distribution in correlation to its concavity, which directly impacts the effectiveness of covert transmission. These findings provide a practical method for determining whether covert communication is achievable based solely on the jammer to adversary channel distribution, offering a clearer and more accessible criterion for covert system design.

Long-duration environmental and physiological monitoring in remote environments requires sensing systems that are compact, energy-efficient, and capable of operating without terrestrial communication infrastructure. This project presents the design and prototype implementation of ResQOS, a dual-device wearable platform that integrates physiological sensing, environmental monitoring, and satellite communication for autonomous emergency detection and remote alert transmission.

The system consists of a Physical Sensing Unit (PSU) worn by the user and an Environmental Sensing Unit (ESU) carried externally. Both devices are built around Nordic nRF52840 microcontrollers and communicate via Bluetooth Low Energy (BLE). The PSU measures physiological parameters including heart rate and blood oxygen saturation (MAX30102), body motion (LIS3DH accelerometer), hydration through bioimpedance sensing, and body temperature. The ESU collects environmental data including temperature, humidity, pressure, and gas concentration (BME688), carbon monoxide (MQ-7), and ambient oxygen (ME2-O2). It also integrates GPS positioning (PA1616D) and a RockBLOCK 9704 modem using the Iridium satellite network for off-grid emergency communication.

Embedded firmware evaluates combined environmental and physiological data to detect hazards such as hypoxia, hypothermia, heat stroke, and dangerous falls. The ESU operates on a 60 s duty cycle, averaging 65 mA and achieving 4–5 days of runtime with a 7500 mAh battery, while the PSU averages 12 mA, enabling 48 hours of operation with a 700 mAh battery. Satellite messages of up to 340 bytes are transmitted with typical 5–20 s latency.

Prototype testing demonstrated stable BLE communication, successful sensor integration, and reliable end-to-end satellite message transmission.

RELATED ABSTRACTS

• Luxury Brands During Recessions, Hynes, Riley Elizabeth, UMass Amherst, Poster Session 5, 3:15 PM - 4:00 PM, Auditorium, A74
• How Does Yoga Benefit an Adolescent Brain and Body?, Almonte, Amy, Northern Essex Community College, Poster Session 3, 1:15 PM - 2:00 PM, Auditorium, A50

Trapped ions are a leading platform for quantum computing, but their performance is currently limited by "anomalous heating", a phenomenon where electric field noise from the trap electrodes destroys the ion's quantum information. While it is known that the distance between the ion and the electrode affects this noise, the impact of complex 3D trap shapes and microscopic surface roughness remains poorly understood. This research develops a novel computational framework to predict how trap shape influences heating rates. Rather than relying solely on experimental measurements, we model the trap interior as a billiard system, simulating how particles would bounce inside the vacuum chamber. By analyzing these trajectories, we investigate whether chaotic orbits correlate with higher noise levels. We employ Topological Data Analysis (TDA), a method from applied mathematics, to quantify the complex structure of these dynamics. The goal is to establish geometric design principles that minimize heating. By identifying which shapes and surface features generate the most noise, this work aims to guide the engineering of next-generation ion traps. This approach bridges the gap between abstract mathematical dynamics and practical quantum hardware, offering a new pathway to more robust and scalable quantum processors.

The integrity of Inertial Measurement Unit (IMU) data is fundamental to the reliability of autonomous systems. This study focuses on acquiring and processing high-fidelity IMU data to support robust multi-modal sensor fusion architectures for platforms operating in environments where Global Navigation Satellite Systems (GNSS) may be unavailable, unreliable, or intentionally disrupted. Achieving reliable navigation under such conditions is critical for next-generation surveillance systems and for meeting stringent anti-radio jamming and spoofing objectives.

Rotor operation in multirotor unmanned aerial vehicles (UAVs) introduces two primary disturbance mechanisms: mechanical vibration that corrupts accelerometer and gyroscope measurements, and electromagnetic interference generated by brushless DC (BLDC) motors and high-current power electronics that distort magnetometer readings. These disturbances scale with rotor throttle and degrade sensor fusion performance. We systematically characterize IMU degradation across controlled throttle sweeps (0–100%) using a PX4-based multirotor platform. Time-domain variance and frequency-domain spectral analyses quantify vibration-induced harmonic amplification, while magnetometer magnitude stability and heading consistency assess throttle-dependent magnetic distortion.

To address these challenges, we propose a data-driven compensation framework centered on an automated data generation pipeline. A cyber-physical testbed leveraging a precision robotic arm induces repeatable interference patterns under controlled conditions representative of GNSS-denied operation. Complementary on-drone sensing modalities, including high-frame-rate optical cameras, are used to infer environmental interference signatures. Machine learning models capture non-linear noise distributions and are integrated into a commercial Extended Kalman Filter (EKF), dynamically refining covariance and state transition parameters. Results demonstrate structured, predictable degradation, enabling adaptive correction strategies that significantly enhance state estimation robustness in high-disturbance environments.

Accurate indoor localization remains a challenging problem due to the limitations of GPS in enclosed environments. Radio-Frequency (RF) techniques offer broad coverage but are sensitive to multipath and environmental variability, while optical wireless systems provide high spatial resolution but suffer from line-of-sight and blockage constraints. This project explores a hybrid RF and optical wireless approach that leverages the complementary strengths of both methods to improve robustness and reliability for indoor localization.

A centralized, multi-node software (SDR) testbed was developed to support synchronized RF and optical data collection across controlled indoor environments. The platform enables repeatable experiments, offline analysis, and future real-time interference by combining distributed sensing nodes with centralized control and processing.

As an initial validation of the testbed and signal quality, raw RF I/Q traces were analyzed offline using a simplified mobility classification task. Time-windowed signal segments were processed in MATLAB and used to distinguish between static and dynamic environmental conditions. This baseline experiment served as a test case to confirm that the collected signals encode meaningful temporal structure and to develop tools applicable to more complex positioning tasks.

Insights gained from the mobility analysis are now guiding the development of RF-optical positioning techniques. By validating the testbed and offline processing pipeline, this work establishes a foundation for evaluating indoor localization methods and transitioning toward real-time, continuous positioning using hybrid wireless signals.

These controlled experiments allow us to validate positioning methods with real measurements rather than simulations, giving confidence in how the system will behave in practical indoor use.

Nanophotonics provides a promising pathway toward scalable quantum information processing by enabling large-scale integration, low optical loss, and room-temperature operation. Among emerging platforms, thin-film lithium niobate (TFLN) has recently gained significant attention due to its strong second-order optical nonlinearity and large electro-optic effect. These properties enable the integration of a universal set of quantum optical operations—including squeezing, modulation, and low-loss routing—on a single chip. This poster presents the design and experimental characterization of several device innovations developed by undergraduate researchers in the Quantum Information Systems Lab at UMass Amherst. Key advances include:

• Fabrication-aware adapted periodic poling for robust quasi-phase matching in non-uniform thin-film waveguides

• Low-loss linear routing components, including novel quasi-Euler fast-quasi-adiabatic (FAQUAD) wavelength multiplexers

• Programmable interferometric circuits for scalable photonic quantum systems

Together, these developments enable improved nonlinear efficiency and ultra-low-loss routing in integrated TFLN quantum photonic circuits, advancing the realization of scalable on-chip quantum technologies.

RELATED ABSTRACTS

• Accelerating Direct Simulation Monte Carlo Methods, Padilla, Francis, UMass Amherst, Poster Session 4, 2:15 PM - 3:00 PM, 163, C13

In the last decade, advances in robotics using artificial intelligence (AI) are starting to make human life easier in both industrial and educational settings. These robots are being trained and used to carry out specific tasks, such as picking up and recognizing objects, and movement. In order to carry out these tasks, they rely on different sensors, such as light, color, and touch sensors. In this project, the main focus is getting the robot to recognize different colored objects, sort them, and implement force-controlled pick-and-place for the colored objects. This shall be executed by programming a Robotic Operating System (ROS) into a Quanser QArm. The QArm is a robotic arm apparatus. It works with MATLAB with Simulink and Python as an implementation platform, along with their add-on QUARC. The building platform can be used to program the ROS. Students can also use their platform for research in Artificial Intelligence (AI) and Machine Learning (ML). This programming can be used with the Quanser QArm to train it to perform specific tasks. These tasks include but are not limited to picking up and putting down objects, and recognizing and sorting objects by other features. The goal is to gain an understanding of how Artificial Intelligence and Machine Learning principles can be applied to servicing industrial and educational settings, as well as demonstrating this understanding by training the Quanser QArm to perform a series of tasks. 

RELATED ABSTRACTS

• Enabling Communication in Multi-Agent VLAs, Benhamou Goldfajn, Dorian, UMass Amherst, Poster Session 3, 1:15 PM - 2:00 PM, 163, C24
• AI in Production Monitoring of Machining Processes, LaValley, Jamesohn, Fitchburg State University, Poster Session 2, 11:30 AM - 12:15 PM, 163, C23
• Bridging the Sim-to-Real Gap: Robust Vision-Based Navigation for Robotic Guide Dogs, Patel, Shiven, UMass Amherst, Poster Session 5, 3:15 PM - 4:00 PM, 165, D4
• Autonomous Photographer, Vasquez, Carlos, Salem State University, Poster Session 2, 11:30 AM - 12:15 PM, Concourse, B7
• Artificial Intelligence in the Operating Room, Patel, Viran, Springfield Technical Community College, Poster Session 2, 11:30 AM - 12:15 PM, Auditorium, A40

RELATED ABSTRACTS

• Design & Evaluation of a Hybrid Energy-Harvesting System for Wearable Health & Environmental Monitoring, Upadhyay, Samanvay, UMass Amherst, Poster Session 4, 2:15 PM - 3:00 PM, Auditorium, A19

RELATED ABSTRACTS

• Alternative Packages through Constraint Perturbation, Booth, Henry Philip, UMass Amherst, Poster Session 3, 1:15 PM - 2:00 PM, 163, C19
• Adaptive Re-Ranking, Genc, Ata Cinar, UMass Amherst, Poster Session 5, 3:15 PM - 4:00 PM, 163, C22