RESEARCH

Research Interests

Construction Robotics; Construction Automation; Human-robot Interaction; Robotic Manipulation Perception Planning and Control; Wearable Physiological Sensing; Flexible/Stretchable Physiological Sensors Design and Fabrication; Immersive Technology for Human-robot Collaboration Training; Multi-agent Robotic Systems; Human-Machine Interaction; 3D Concrete Printing; Engineering Education

Future of Construction Workplace Health Monitoring

Despite the increasing number of work-related injuries and illnesses among construction workers, there is a lack of efficient health and safety monitoring methods

at construction jobsites. The current methods for evaluating workers’ physical and mental health mostly rely on self-assessment measures with underlying challenges (e.g., being subjective and intrusive) for application in the field.

While the advent of wearable physiological sensing and artificial intelligence (AI) has provided opportunities to objectively and non-invasively evaluate workers’ health parameters, several fundamental challenges still need to be addressed before a health monitoring approach for construction jobsites can be established. This project aims to develop a worker-centered health and safety monitoring system that integrates advancements in wearable biosensor design, physiological sensing, predictive analysis, and construction safety to transform the health and safety monitoring of the construction workforce.

The objective of this research is achieved through designing and fabricating advanced wearable biosensors, developing machine learning algorithms robust to inter-and intra-individual variability, and visualizing health and safety information of construction sites through an array of collective health analyses represented through health and safety maps. This project will lay the groundwork for enhanced proactive health and safety monitoring in construction sites using the collective visualization of workers’ health and safety information.

Publications (selected):

Ojha, A., Shakerian, S., Habibnezhad, M., and Jebelli, H. (2023). "Feasibility Verification of Multimodal Wearable Sensing System for Holistic Health Monitoring of Construction Workers," Proceedings of the 2021 Canadian Society for Civil Engineering (CSCE), Montreal, Quebec.
Liu, Y., Gautam, Y., Shayesteh, S., Jebelli, H., and Mahdi Khalili, M. (2024). "Towards an Efficient Physiological-Based Worker Health Monitoring System in Construction: An Adaptive Filtering Method for Removing Motion Artifacts in Physiological Signals of Workers," 2023 International Conference on Computing in Civil Engineering (i3CE), Corvallis, Oregon, U.S.
Shayesteh, S., and Jebelli, H. (2024). "Evaluating the Feasibility of Personalized Health Status Feedback to Enhance Worker Safety and Well-Being at Construction Jobsites." In Construction Research Congress 2024 (pp. 575-585), Des Moines, Iowa, U.S.
Gautam, Y., Liu, Y., and Jebelli, H. (2024). "Unsupervised Adversarial Domain Adaptation in Wearable Physiological Sensing for Construction Workers’ Health Monitoring Using Photoplethysmography." In Construction Research Congress 2024 (pp. 339-348), Des Moines, Iowa, U.S.

Funding:

National Science Foundation (NSF) - Future of Work at the Human-Technology Frontier program
Role: lead PI; Award Amount: $1.8m Start Date: November 27, 2023

Featured News:

UIUC News - Advanced health monitoring of construction workers is focus of newly funded project

Worker-Centered Human-Robot Partnership

To address the long-standing challenges in the construction industry (e.g., stagnant productivity and labor shortages), cobots are emerging at construction sites. Cobots that operate with human assistance, such as bricklaying robots, are designed to integrate the tirelessness and repeatability of automatons with the flexibility and decision-making ability of human workers. However, in any industrial sector, the adoption of new technologies requires careful monitoring and consideration. Robots at construction sites are no exception. While cobots have shown great potential in

improving stagnant productivity and labor shortages, due to the lack of perception and reasoning capabilities, the robot cannot understand its co-workers' positions and motion information. When workers move in close proximity to cobots, the interaction between cobots and workers may trigger collision accidents. In addition, cobots can cause high levels of physical fatigue and excessive cognitive load to their co-workers due to unparalleled work performance. In the long run, prolonged physical fatigue and cognitive load can increase workers' physical injury, anxiety, and emotional distress. To ensure the physical, physiological, and psychological safety of construction workers during HRC, this project aims to develop a series of worker-centered human-robot co-adaptation mechanisms by innovating and applying computer vision, artificial intelligence, physiological sensing, robot perception, and robot behavior adaptation techniques. The proposed mechanisms will enable cobots to reason about the next actions of the workers; measure the probability of collisions with workers; assess the physiological and psychological states of workers; and generate optimal robot adjustments to minimize the physical and psychological safety risks of workers. This interdisciplinary research can facilitate the safe and efficient implementation of collaborative robots in the construction industry.

Publications (selected):

Liu, Y., Habibnezhad, and Jebelli, H. (2021). "Brainwave-driven human-robot collaboration in construction" Automation in Construction, Elsevier, 124, 103556. https://doi.org/10.1016/j.autcon.2021.103556.
Liu, Y., Habibnezhad, M., and Jebelli, H. (2021) "Worker-aware Task Planning for Construction Robots: A Physiologically Based Communication Channel Interface," In: Jebelli, H., Habibnezhad, M., Shayesteh, S., Asadi. S., Lee. S., (eds) Automation and Robotics in the Architecture, Engineering, and Construction Industry, Springer, Cham. https://doi.org/10.1007/978-3-030-77163-8_9.
Liu, Y., Ojha, A., Shayesteh, S., and Jebelli, H. (2022). "Robotic Sensing and Perception in Construction: Generative Adversarial Networks (GANs) based Physiological Computing Mechanism to Enable Robots to Perceive Workers' Cognitive Load" Canadian Journal of Civil Engineering, Canadian Science Publishing. http://doi.org/10.1139/cjce-2021-0646.
Liu, Y., and Jebelli, H. (2023). "Intention‐aware robot motion planning for safe worker–robot collaboration," Computer‐Aided Civil and Infrastructure Engineering. https://doi.org/10.1111/mice.13129

Featured News:

“These Robots Want to Read Your Mind While You Work – You Should Let Them, Inverse Magazine, New York, NY, U.S. Link: https://www.inverse.com/innovation/mind-reading-robots-are-the-future-of-work
“Human-Robot Collaboration in Construction,” Penn State Office of Strategic Communications, Penn State News, The Pennsylvania State University, State College, PA, U.S. Link: https://news.engr.psu.edu/2021/jebelli-houtan-robots-construction.aspx

Wearable Assistive Robotics for Field Workers

A wide range of wearable robots is emerging as solutions to human operation challenges in the construction sector. The aim of such

solutions is to reduce work-related physical injuries among workers by providing lift support, weight dispersion, and posture correction. In spite of the potential of these wearable robots to reduce the physical demands of construction workers, the

current body of knowledge does not provide an adequate understanding of the risks and challenges of incorporating these robots on construction sites. By understanding the potential physical, psychological, and socio-technical risks of these wearable robots at construction sites, this project aims to overcome the challenges of scalable adoption of these wearable robots in the construction industry, thereby improving the safety and productivity of 7.5 million workers in the U.S. construction sector. Furthermore, this research will provide empirical evidence for manufacturers to design more adaptable, accessible, acceptable, and comfortable wearable robots for a wider range of body shapes and sizes to take into account the diverse populations of the construction sector. This interdisciplinary research seeks to integrate advances across a diverse spectrum of critical innovations, including immersive technologies, physiological sensing, wearable robots, and organizational psychology, to identify the underlying physical, psychological, and socio-technical risks of exoskeletons in the construction sector.

Publications (selected):

Ojha, A., Guo, H., Jebelli, H., Martin, A., and Akanmu, A. (2024). "Assessing the Impact of Active Back Support Exoskeletons on Muscular Activity during Construction Tasks: Insights from Physiological Sensing," 2023 International Conference on Computing in Civil Engineering (i3CE), Corvallis, Oregon, U.S.
Breneman, M., Ojha, A., Jebelli, H., Simkins, S. J., and Akanmu, A. (2024). "Breaking Down Barriers: A Study of Challenges to Adopting Powered Exoskeletons in the US Construction Industry," 2023 International Conference on Computing in Civil Engineering (i3CE), Corvallis, Oregon, U.S.
Liu, Y., Gautam, Y., Ojha, A., Shayesteh, S., and Jebelli, H. (2024). "Studying the Effects of Back-Support Exoskeletons on Workers’ Cognitive Load during Material Handling Tasks." In Construction Research Congress 2024 (pp. 659-669), Des Moines, Iowa, U.S.

Funding:

National Science Foundation (NSF) -- National Robotics Initiative 3.0 (NRI) Program
Role: lead PI; Award Amount: $675K; Start Date: January 17, 2024

Featured News:

PennState News – One robot-assisted step closer to construction safety with NSF grant
UIUC News - NSF National Robotics Initiative Grant Aims to Revolutionize Construction Sites with Wearable Robots

Immersive VR-based Learning Environment for Learning Human-Robot Collaboration

As the construction industry gears towards widespread acceptance of

robots, there is a need to prepare Construction Engineering and Management (CEM) students and the future workforce to excel in highly technological working environments. However, the current curriculum is inadequate to provide individuals with the required competencies for working with construction robots. To fill this gap, this project aims to develop a personalized, immersive learning environment for human-robot collaboration in construction.

The proposed research integrates advances in construction engineering, field robotics, learning sciences, educational psychology, and human-computer interaction to improve our understanding of the challenges that individuals face in learning how to safely interact and collaborate with robots during construction work. This work will establish a competency framework to define the knowledge, skills, and abilities required of the future workforce to interact with robots in the construction industry. The development of the immersive learning environment will help us investigate ways in which domain-specific knowledge can be embedded in VR environments.

Publications (selected):

Shayesteh, S., Ojha, A., Liu, Y., and Jebelli, H. (2024). "Evaluating the Impact of Powered Back-Support Exoskeletons and Virtual Reality Interventions on Gait Stability of Construction Workers at Heights," 2023 International Conference on Computing in Civil Engineering (i3CE), Corvallis, Oregon, U.S.
Ojha, A., Shayesteh, S., Jebelli, H., Akanmu, A., Miller, S., and Sharma, P. (2024). "Revitalizing Students’ Perception and Knowledge of Construction Safety: Leveraging Virtual Reality as an Experiential Learning Tool for Construction Robotic Safety Education." In Construction Research Congress 2024 (pp. 129-138), Des Moines, Iowa, U.S.

Funding:

National Science Foundation (NSF) - Improving Undergraduate STEM Education (IUSE) Program
Role: lead PI; Award Amount: $600K; Start Date: January 8, 2024

Featured News:

UIUC News - New Grant Fuels Development of Innovative VR-Based Robotics Training Tool

Human-Centered Heat Stress Risk Assessment

The construction industry exposes many workers to a heat-vulnerable environment with limited sets of fundamental precautions against heat stress. It leads to increased chances of developing cardiovascular, respiratory, and chronic kidney diseases in workers. In case of severe heat stress, it can even result in immediate effects such as organ failure or death in some instances. Unfortunately, current heat stress regulation practices lack a detailed protocol for early heat stress prediction to intervene before the heat

stress level becomes critical. This is because the response of bodies toward heat exposure varies highly among individuals, and the current heat-stress measurement techniques are not adequately suitable for monitoring workers’ heat-stress exposure in the field as they are nonviable, invasive, imprecise, and discontinuous. Realizing this huge gap, we are developing a non-invasive sensing system and generating a data-driven framework that can access, understand, and predict heat stress exposure at job sites in near real time. This system will protect workers from severe heat-related injuries by triggering safety feedback to workers timely. The system will also allow new workers to safely start their job, build up a tolerance for hot conditions, and acclimatize to the workplace.

Publications (selected):

Ojha, A., Shakerian, S., Habibnezhad, M., Jebelli, H., Lee, S., Fardhosseini, M.S. (2020). “Feasibility of Using Physiological Signals from a Wearable Biosensor to Monitor Dehydration of Construction Workers,” Creative Construction e-Conference 2020, Opatija, Croatia.
S Shakerian, M Habibnezhad, A Ojha, G Lee, Y Liu, H Jebelli, SH Lee (2021). “Assessing occupational risk of heat stress at construction: A worker-centric wearable sensor-based approach,” Journal of Safety Science, 142, 10.1016/j.ssci.2021.105395.
Ojha, A., Sharifironizi, A., Liu, Y., and Jebelli, H. (2024). "Enhancing Human-Centric Physiological Data-Driven Heat Stress Assessment in Construction through a Transfer Learning-Based Approach." In Construction Research Congress 2024 (pp. 157-167), Des Moines, Iowa, U.S.
Ojha, A., Jebelli, H., Alexander, L., and Loeffert, J. R. (2024). "Quantifying the Implications of Humidity and Temperature on Heat Stress Exposure of Construction Workers: A Worker-Centric Physiological Sensing Approach." In Construction Research Congress 2024 (pp. 196-205), Des Moines, Iowa, U.S.

Funding:

National Institute for Occupational Safety and Health (NIOSH)
Role: lead PI; Award Amount: $425K; Start Date: September 3, 2023

Featured News:

PennState News – Researchers to address heat-related illnesses on construction sites
UIUC News - Innovative AI-driven framework with flexible biosensors aims to combat worker heat stress

Human-Robot Teaming Training in Construction

Current trends in the construction sector show an increased emphasis on utilizing robotic technologies across the industry to address long-lasting challenges, such as the aging workforce, workforce shortages, and stagnant productivity. As the industry moves toward the widespread implementation of robots, there is a need to prepare the construction workforce to develop the required knowledge and skills to succeed in increasingly technological work environments. However, most of the current training programs in construction are inadequate to equip workers and engineers with the required multi-disciplinary skills to

work alongside construction robots. This deficiency mostly stems from the lack of a clear understanding of the required competencies to work with robots and the lack of an effective learning environment to provide the trainees with hands-on learning experiences.

To fill these gaps, this project aims to investigate a virtual reality-based training environment for developing the required hands-on skills of interacting with robots in construction jobsites. The development of an immersive learning environment enables us to integrate interactive training modules with various life-like scenarios and investigate how domain-specific knowledge can be embedded in such scenarios. To that end, virtual representations of the trainee (i.e., virtual avatar) and physiological sensing mechanisms are utilized to deliver a tailored learning experience and evaluate the trainee's performance throughout various human-robot teaming scenarios. The results of this project can pave the way for developing more effective training practices to enhance workers’ safety during human-robot interactions.

Publications (selected):

Shayesteh, S., and Jebelli, H. (2021). "Feasibility of Virtual Avatar Simulator for Human-robot Collaboration Training in Construction," 2021 International Conference on Computing in Civil Engineering (I3CE 2021), Orlando, FL, U.S.
Shayesteh, S., and Jebelli, H. (2022). “Enhanced Situational Awareness in Worker-Robot Interaction in Construction: Assessing the Role of Visual Cues,” Construction Research Congress 2022 (ASCE), Arlington, Virginia, U.S.
Shayesteh, S., Ojha, A., Liu, Y., and Jebelli, H. (2023). "Human-robot teaming in construction: Evaluative safety training through the integration of immersive technologies and wearable physiological sensing," Safety Science, Elsevier, 159. https://doi.org/10.1016/j.ssci.2022.106019

Funding:

The Leonhard Center for Enhancement of Engineering Education
Role: PI; Start Date: June 2021

Infusion of Data Science and Computation into Engineering Curricula

This project aims to develop a curricular framework for data science education and workforce development that is transferable between diverse institutions.

RAISe lab will focus on robotics and automation curriculum development and implementation. The objectives of this project are to (1) facilitate data science education and workforce development for engineering and related topics, (2) provide

opportunities for students to participate in practical experiences where they can learn new skills in a variety of environments, and (3) expand the data science talent pool by enabling the participation of undergraduate students with diverse backgrounds, experiences, skills, and technical maturity in the Data Science Corps. This work will support the Data Science Corps objective of building capacity for education and workforce development to harness the data revolution at local, state, and national levels.

Funding:

National Science Foundation (NSF) - Harnessing the Data Revolution: Data Science Corps program
Role: Co-PI; Award Amount: $1.3m; Start Date: October 2021

Featured News:

PennState News – Impacting students and the community

Digital Twin Enabled Facilities Maintenance

Facility maintenance activities account for 40% of total building operational costs, which can be significantly reduced by optimizing the frequency of maintenance activities. Traditional approaches use scheduled strategies that tend to over-maintain the facility, increasing the cost and time associated with maintenance. Thus, an optimal maintenance strategy is required such that the possible breakdowns in HVAC installations can be predicted early on,

and maintenance can be carried out right before the actual breakdown. Digital Twin is the state-of-art approach that allows continuous monitoring and prediction of system behavior by utilizing the current advancement in sensor technologies, IoT, and computation. At RAISE Lab, we have developed a data-driven digital twin of an HVAC system that utilizes advanced machine learning techniques and IoT platforms in a facility monitoring and prediction framework to establish near-real-time communication between physical and digital systems. The digital twin is capable of alerting facility managers about anticipated failures on time to allow dynamic decision-making in facility maintenance management.

Publications (selected):

Dissertation: "Digital Twin Enabled Facilities Maintenance Management through the Integration of Artificial Intelligence and Sensory-Level Data"; https://etda.libraries.psu.edu/catalog/22115sqs6689
Shakerian, S., Ojha, A., Jebelli, H., and Sitzabee, W., (2021). "Investigating the Potentials of Operational Data Collected from Facilities' Embedded Sensors for Early Detection of HVAC Systems' Failures," 2021 International Conference on Computing in Civil Engineering (i3CE), Orlando, Florida, U.S.
Shakerian, S., Jebelli, H., and Sitzabee, W., (2021). "Improving the Prediction Accuracy of Data-Driven Fault Diagnosis for HVAC Systems by Applying Synthetic Minority Oversampling Technique," 2021 International Conference on Computing in Civil Engineering (i3CE), Orlando, Florida, U.S.

Funding:

Penn State's Office of Physical Plant

Human-Cyber-Physical Framework for Occupant-Centric Control of Building Facilities

The building sector represents around 30% of global energy use, and 28% of global CO2 emissions, with more than 40% projected growth in the next 20 years. This reflects the urgency to minimize the energy consumption of our built environment. Occupant behavior has been identified as a major factor influencing energy consumption during its operation. However, current building design and operation practices fail to integrate occupant-related

information leading to non-optimal building operations. In addition, the buildings often fail to meet occupant satisfaction due to the lack of contextual awareness in the building facility control system. Acknowledging these issues, this project aims to develop a Human-Cyber-Physical framework for the Occupant-Centric Control (OCC) of building facilities.

OCC is a control strategy to achieve contextually aware building operation in which occupant-related information is used to train models that can predict the occupants’ needs in a built environment and adjust the performance of different building facilities accordingly. To implement this strategy, we are developing a framework using advanced machine-learning techniques and predictive control methods, which establishes continuous communication between the occupants (human), the digital model (cyber), and the physical system (physical). Within this control framework, the system can predict and respond to the changes in occupants’ behavior automatically to ensure optimal energy consumption and occupant satisfaction in all conditions. This work will contribute to the development of smart building technologies for the zero-carbon future of the building energy sector.

Publications (selected):

Adhikari, R., Gautam, Y., Jebelli, H., and Sitzabee, W. E. (2024). "Deep Learning and Reinforcement Learning for Modeling Occupants’ Information in an Occupant-Centric Building Control: A Systematic Literature Review." In Construction Research Congress 2024 (pp. 186-195), Des Moines, Iowa, U.S.

Funding:

Penn State's Office of Physical Plant