Amna Mazen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Amna Mazen has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Mohan Krishnan
Department of Electrical and Computer Engineering and Computer Science, University of Detroit Mercy, Detroit, MI, USA
5 shared publications - Mariam Faied
University of Detroit Mercy, Detroit, MI, USA
4 shared publications - Ana Dyreson
Michigan Technological University
2 shared publications - Ashraf Saleem
Department of Applied Computing, College of Computing, Michigan Technological University, Houghton, MI, USA
2 shared publications Brendan McClanahan
Department of Mechanical Engineering, University of Detroit Mercy, Detroit, USA
2 shared publicationsJonathan M. Weaver
Department of Mechanical Engineering, University of Detroit Mercy, Detroit, USA
2 shared publications
Latest Publications
- Open Access
An Autonomous Robotic System for Object Retrieval and Delivery: Enhancing Independence for Users Living with Disability and Older Adults
As the global population ages, there is a growing need for assistive technologies to help older adults maintain their independence. This work presents a cost-effective autonomous socially assistive robot designed for object retrieval and delivery, enhancing accessibility in home environments. The system is built on the Robot Operating System (ROS) framework and integrates three key components: the Pioneer P3-DX mobile robot for autonomous navigation, the ReactorX-200 robotic arm for pick-and-place operations, and the Kinect v2 RGB-D camera for object detection and localization. Users interact with the robot through natural language processing by issuing voice commands to retrieve various objects. Microsoft Azure-powered speech recognition processes these commands to extract keywords and then localize requested objects on a predefined building map. Pioneer P3-DX, equipped with a Hokuyo LiDAR, enables autonomous navigation and obstacle avoidance, while Kinect v2, integrated with the YOLOv8 algorithm, facilitates object recognition and localization. The robot retrieves and delivers the user’s requested objects while following the shortest available path. Experimental evaluations in a home environment demonstrate the system’s effectiveness in identifying and retrieving requested objects. The subsystems achieve a success rate of 85–95% across more than 50 runs, highlighting their strong performance. The proposed approach provides a proof of concept for future advancements in assistive robotics, demonstrating the seamless integration of advanced technologies into a cost-effective and user-friendly platform.
Open AccessAn Autonomous Robotic System for Object Retrieval and Delivery: Enhancing Independence for Users Living with Disability and Older Adults
As the global population ages, there is a growing need for assistive technologies to help older adults maintain their independence. This work presents a cost-effective autonomous socially assistive robot designed for object retrieval and delivery, enhancing accessibility in home environments. The system is built on the Robot Operating System (ROS) framework and integrates three key components: the Pioneer P3-DX mobile robot for autonomous navigation, the ReactorX-200 robotic arm for pick-and-place operations, and the Kinect v2 RGB-D camera for object detection and localization. Users interact with the robot through natural language processing by issuing voice commands to retrieve various objects. Microsoft Azure-powered speech recognition processes these commands to extract keywords and then localize requested objects on a predefined building map. Pioneer P3-DX, equipped with a Hokuyo LiDAR, enables autonomous navigation and obstacle avoidance, while Kinect v2, integrated with the YOLOv8 algorithm, facilitates object recognition and localization. The robot retrieves and delivers the user’s requested objects while following the shortest available path. Experimental evaluations in a home environment demonstrate the system’s effectiveness in identifying and retrieving requested objects. The subsystems achieve a success rate of 85–95% across more than 50 runs, highlighting their strong performance. The proposed approach provides a proof of concept for future advancements in assistive robotics, demonstrating the seamless integration of advanced technologies into a cost-effective and user-friendly platform.
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 34 (3) , 1248-1258
Bézier Curve-Based Optimization Framework for Multirobot Motion Planning in Unknown Environments
In recent years, multirobot systems (MRSs) have witnessed significant growth because of their capability of solving complex problems involving collaborative missions. A two-phase approach is typically employed in such applications, encompassing task planning and assignment among the robots, followed by motion planning to execute the assigned tasks. This work focuses on the motion planning phase, which facilitates the optimal navigation of multiple robots from their initial to desired goal configurations. A novel multirobot Bézier curve-based trajectory planner (MrBTP) is proposed. MrBTP generates optimal trajectories for multiple robots navigating in an unknown environment, including the polygonal characterization of the available open space. The optimization process is based on using a weighted objective function that combines the navigation time of each robot and the length of its trajectory, allowing for various path optimization strategies. The effectiveness of MrBTP has been evaluated in various scenarios, and its performance is compared with that of a selected trajectory planner based on three performance metrics. Over the three robots used in the test, MrBTP reduced the navigation time by 21%–49% and the trajectory length by 10%–18% compared to the selected benchmark trajectory planner. In addition, it ensures $C^{2}$ rather than just $C^{1}$ continuity.
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 34 (3) , 1248-1258
Bézier Curve-Based Optimization Framework for Multirobot Motion Planning in Unknown Environments
In recent years, multirobot systems (MRSs) have witnessed significant growth because of their capability of solving complex problems involving collaborative missions. A two-phase approach is typically employed in such applications, encompassing task planning and assignment among the robots, followed by motion planning to execute the assigned tasks. This work focuses on the motion planning phase, which facilitates the optimal navigation of multiple robots from their initial to desired goal configurations. A novel multirobot Bézier curve-based trajectory planner (MrBTP) is proposed. MrBTP generates optimal trajectories for multiple robots navigating in an unknown environment, including the polygonal characterization of the available open space. The optimization process is based on using a weighted objective function that combines the navigation time of each robot and the length of its trajectory, allowing for various path optimization strategies. The effectiveness of MrBTP has been evaluated in various scenarios, and its performance is compared with that of a selected trajectory planner based on three performance metrics. Over the three robots used in the test, MrBTP reduced the navigation time by 21%–49% and the trajectory length by 10%–18% compared to the selected benchmark trajectory planner. In addition, it ensures $C^{2}$ rather than just $C^{1}$ continuity.
- Open Access
Overcoming Variable Illumination in Photovoltaic Snow Monitoring: A Real-Time Robust Drone-Based Deep Learning Approach
Snow accumulation on photovoltaic (PV) panels can cause significant energy losses in cold climates. While drone-based monitoring offers a scalable solution, real-world challenges like varying illumination can hinder accurate snow detection. We previously developed a YOLO-based drone system for snow coverage detection using a Fixed Thresholding segmentation method to discriminate snow from the solar panel; however, it struggled in challenging lighting conditions. This work addresses those limitations by presenting a reliable drone-based system to accurately estimate the Snow Coverage Percentage (SCP) over PV panels. The system combines a lightweight YOLOv11n-seg deep learning model for panel detection with an adaptive image processing algorithm for snow segmentation. We benchmarked several segmentation models, including MASK R-CNN and the state-of-the-art SAM2 segmentation model. YOLOv11n-seg was selected for its optimal balance of speed and accuracy, achieving 0.99 precision and 0.80 recall. To overcome the unreliability of static thresholding under changing lighting, various dynamic methods were evaluated. Otsu’s algorithm proved most effective, reducing the absolute error of the mean in SCP estimation to just 1.1%, a significant improvement over the 13.78% error from the previous fixed-thresholding approach. The integrated system was successfully validated for real-time performance on live drone video streams, demonstrating a highly accurate and scalable solution for autonomous snow monitoring on PV systems.
Open AccessOvercoming Variable Illumination in Photovoltaic Snow Monitoring: A Real-Time Robust Drone-Based Deep Learning Approach
Snow accumulation on photovoltaic (PV) panels can cause significant energy losses in cold climates. While drone-based monitoring offers a scalable solution, real-world challenges like varying illumination can hinder accurate snow detection. We previously developed a YOLO-based drone system for snow coverage detection using a Fixed Thresholding segmentation method to discriminate snow from the solar panel; however, it struggled in challenging lighting conditions. This work addresses those limitations by presenting a reliable drone-based system to accurately estimate the Snow Coverage Percentage (SCP) over PV panels. The system combines a lightweight YOLOv11n-seg deep learning model for panel detection with an adaptive image processing algorithm for snow segmentation. We benchmarked several segmentation models, including MASK R-CNN and the state-of-the-art SAM2 segmentation model. YOLOv11n-seg was selected for its optimal balance of speed and accuracy, achieving 0.99 precision and 0.80 recall. To overcome the unreliability of static thresholding under changing lighting, various dynamic methods were evaluated. Otsu’s algorithm proved most effective, reducing the absolute error of the mean in SCP estimation to just 1.1%, a significant improvement over the 13.78% error from the previous fixed-thresholding approach. The integrated system was successfully validated for real-time performance on live drone video streams, demonstrating a highly accurate and scalable solution for autonomous snow monitoring on PV systems.
- Open Access
Estimating Snow Coverage Percentage on Solar Panels Using Drone Imagery and Machine Learning for Enhanced Energy Efficiency
Snow accumulation on solar panels presents a significant challenge to energy generation in snowy regions, reducing the efficiency of solar photovoltaic (PV) systems and impacting economic viability. While prior studies have explored snow detection using fixed-camera setups, these methods suffer from scalability limitations, stationary viewpoints, and the need for reference images. This study introduces an automated deep-learning framework that leverages drone-captured imagery to detect and quantify snow coverage on solar panels, aiming to enhance power forecasting and optimize snow removal strategies in winter conditions. We developed and evaluated two approaches using YOLO-based models: Approach 1, a high-precision method utilizing a two-class detection model, and Approach 2, a real-time single-class detection model optimized for fast inference. While Approach 1 demonstrated superior accuracy, achieving an overall precision of 89% and recall of 82%, it is computationally expensive, making it more suitable for strategic decision making. Approach 2, with a precision of 93% and a recall of 75%, provides a lightweight and efficient alternative for real-time monitoring but is sensitive to lighting variations. The proposed framework calculates snow coverage percentages (SCP) to support snow removal planning, minimize downtime, and optimize power generation. Compared to fixed-camera-based snow detection models, our approach leverages drone imagery to improve detection precision while offering greater scalability to be adopted for large solar farms. Qualitative and quantitative analysis of both approaches is presented in this paper, highlighting their strengths and weaknesses in different environmental conditions.
Open AccessEstimating Snow Coverage Percentage on Solar Panels Using Drone Imagery and Machine Learning for Enhanced Energy Efficiency
Snow accumulation on solar panels presents a significant challenge to energy generation in snowy regions, reducing the efficiency of solar photovoltaic (PV) systems and impacting economic viability. While prior studies have explored snow detection using fixed-camera setups, these methods suffer from scalability limitations, stationary viewpoints, and the need for reference images. This study introduces an automated deep-learning framework that leverages drone-captured imagery to detect and quantify snow coverage on solar panels, aiming to enhance power forecasting and optimize snow removal strategies in winter conditions. We developed and evaluated two approaches using YOLO-based models: Approach 1, a high-precision method utilizing a two-class detection model, and Approach 2, a real-time single-class detection model optimized for fast inference. While Approach 1 demonstrated superior accuracy, achieving an overall precision of 89% and recall of 82%, it is computationally expensive, making it more suitable for strategic decision making. Approach 2, with a precision of 93% and a recall of 75%, provides a lightweight and efficient alternative for real-time monitoring but is sensitive to lighting variations. The proposed framework calculates snow coverage percentages (SCP) to support snow removal planning, minimize downtime, and optimize power generation. Compared to fixed-camera-based snow detection models, our approach leverages drone imagery to improve detection precision while offering greater scalability to be adopted for large solar farms. Qualitative and quantitative analysis of both approaches is presented in this paper, highlighting their strengths and weaknesses in different environmental conditions.
- 3 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Intelligent Warehouse Robot for Classifying, Sorting, and Transporting Goods’ Boxes
With the rapid development of global retail e-commerce, the critical role of warehouses in logistics systems has become increasingly prominent. Current state-of-the-art solutions typically address only one or two warehouse tasks, including object recognition, sorting, and transportation. This paper presents the development of an integrated autonomous warehouse robot system capable of performing the three functions. The system mimics human capabilities by integrating various components, including advanced computer vision technologies using YOLOv5, a mobile robot, and a robotic arm. The YOLOv5 algorithm is employed for object detection and size estimation, enabling the system to classify goods-containing boxes into small, medium, and large categories. These classified goods are then delivered to predetermined locations by the mobile robot. Finally, the robot arm is employed to pick and place boxes in their storage location based on size. Experimental results demonstrate the system’s robustness and reliability, highlighting its significant potential to enhance warehouse management through robotic automation.
- 3 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Intelligent Warehouse Robot for Classifying, Sorting, and Transporting Goods’ Boxes
With the rapid development of global retail e-commerce, the critical role of warehouses in logistics systems has become increasingly prominent. Current state-of-the-art solutions typically address only one or two warehouse tasks, including object recognition, sorting, and transportation. This paper presents the development of an integrated autonomous warehouse robot system capable of performing the three functions. The system mimics human capabilities by integrating various components, including advanced computer vision technologies using YOLOv5, a mobile robot, and a robotic arm. The YOLOv5 algorithm is employed for object detection and size estimation, enabling the system to classify goods-containing boxes into small, medium, and large categories. These classified goods are then delivered to predetermined locations by the mobile robot. Finally, the robot arm is employed to pick and place boxes in their storage location based on size. Experimental results demonstrate the system’s robustness and reliability, highlighting its significant potential to enhance warehouse management through robotic automation.
- 17 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 570-575
An Interactive Robotic System for Efficient Book Borrowing and Returning in a Library Environment
Navigating a large library can be an overwhelming task for many users. The challenge of locating specific books often results in wasted time and diminished user experience. This work introduces an interactive robotic system to assist library users in borrowing and returning books. The system utilizes facial recognition technology to identify and greet users upon entry into its detection range. For book borrowing, a speech recognition algorithm queries users about the books’ titles they wish to borrow. It then accesses a database modeled to mimic the University of Detroit Mercy Library database. Subsequently, the robot guides users to retrieve the requested books from their corresponding bookshelves on the library map using the shortest possible path. After completing the retrieving task, the robot autonomously returns to its initial greeting position within the library. For book returns, a QR code algorithm enables users to scan the QR code on the book cover in front of the robot’s camera. The borrowing and returning procedures have been tested in real-time under various scenarios to ensure reliability and effectiveness. Furthermore, an interactive library website has been developed to assist users in checking the availability and precise location of books online. This website features distinct login privileges for students and administrators, facilitating access to essential library information and services.
- 17 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 570-575
An Interactive Robotic System for Efficient Book Borrowing and Returning in a Library Environment
Navigating a large library can be an overwhelming task for many users. The challenge of locating specific books often results in wasted time and diminished user experience. This work introduces an interactive robotic system to assist library users in borrowing and returning books. The system utilizes facial recognition technology to identify and greet users upon entry into its detection range. For book borrowing, a speech recognition algorithm queries users about the books’ titles they wish to borrow. It then accesses a database modeled to mimic the University of Detroit Mercy Library database. Subsequently, the robot guides users to retrieve the requested books from their corresponding bookshelves on the library map using the shortest possible path. After completing the retrieving task, the robot autonomously returns to its initial greeting position within the library. For book returns, a QR code algorithm enables users to scan the QR code on the book cover in front of the robot’s camera. The borrowing and returning procedures have been tested in real-time under various scenarios to ensure reliability and effectiveness. Furthermore, an interactive library website has been developed to assist users in checking the availability and precise location of books online. This website features distinct login privileges for students and administrators, facilitating access to essential library information and services.
- Open Access
- 11 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 8655-8667
Optimal Kinodynamic Trajectory Planner for Mobile Robots in an Unknown Environment Using Bézier Contours
Trajectory planning in the field of mobile robotics involves the generation of a trajectory to navigate a robot from a start state to a goal state. One widely employed technique involves a two-step approach: a path planner generates a path made up of piecewise linear segments with sharp turns, which are then smoothed in the trajectory generation step. In contrast, this work formulates trajectory generation as an optimization problem based on the Bézier curve, denoted as ‘BTP’, to generate the robot’s trajectory in one step. It uses a weighted objective function of trajectory length and navigation time to suit different optimization strategies while considering the robot’s kinematics and dynamics limitations. BTP adopts matrix-based formulations for all mathematical operations to enable dynamic adjustment of the degree of the Bézier curve during the optimization process, if convergence is not obtained with the current degree. Additionally, BTP guarantees that the robot’s trajectory is always within the open space identified by the robot’s sensors. The efficacy of BTP has been evaluated through simulations and real-world experimentation, including soccer games and cluttered environment scenarios. Finally, the performance is benchmarked against some of the existing trajectory planners. BTP reduced the robot’s navigation time by a minimum of 11% up to 55% compared to other tested trajectory planners, ensuring $C^{2}$ continuity rather than just $C^{1}$ continuity. Furthermore, it consistently achieved precise goal configuration, unlike the tested trajectory planners, which exhibited deviations of up to 0.6 meters.
- 11 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 8655-8667
Open AccessOptimal Kinodynamic Trajectory Planner for Mobile Robots in an Unknown Environment Using Bézier Contours
Trajectory planning in the field of mobile robotics involves the generation of a trajectory to navigate a robot from a start state to a goal state. One widely employed technique involves a two-step approach: a path planner generates a path made up of piecewise linear segments with sharp turns, which are then smoothed in the trajectory generation step. In contrast, this work formulates trajectory generation as an optimization problem based on the Bézier curve, denoted as ‘BTP’, to generate the robot’s trajectory in one step. It uses a weighted objective function of trajectory length and navigation time to suit different optimization strategies while considering the robot’s kinematics and dynamics limitations. BTP adopts matrix-based formulations for all mathematical operations to enable dynamic adjustment of the degree of the Bézier curve during the optimization process, if convergence is not obtained with the current degree. Additionally, BTP guarantees that the robot’s trajectory is always within the open space identified by the robot’s sensors. The efficacy of BTP has been evaluated through simulations and real-world experimentation, including soccer games and cluttered environment scenarios. Finally, the performance is benchmarked against some of the existing trajectory planners. BTP reduced the robot’s navigation time by a minimum of 11% up to 55% compared to other tested trajectory planners, ensuring $C^{2}$ continuity rather than just $C^{1}$ continuity. Furthermore, it consistently achieved precise goal configuration, unlike the tested trajectory planners, which exhibited deviations of up to 0.6 meters.
- 1 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1754-1760
Occupancy Detection based on WI-FI SysLog Data
Residential and commercial buildings consume almost 27% of the total energy in the United States. Therefore, accurate occupancy detection is necessary and valuable to increase the efficiency of the building’s Heating, Ventilation, and Air Conditioning (HVAC) system while maintaining the occupants’ comfort. The occupancy detection process involves deploying two fundamental types of sensors - environmental sensors (humidity, carbon dioxide, temperature, and light sensors) and specialized sensors (Passive infrared, Radio Frequency Identification sensors, cameras, and WI-FI). These sensors are used with three primary categories of algorithms – analytical, knowledge-based, and data-driven. Our paper uses Syslog data from a university building network to generate eight different datasets after removing non-essential data and retaining only those needed. A framework is then proposed, which consists of two stages. The first stage compares the results of four machine learning algorithms, including K-Nearest Neighbors (KNN), Random Forest (RF), eXtreme Gradient Boosting (XGB), and Logistic Regression (LR), to choose the best two models. In this work, XGB and RF produce the most accurate results. The second stage tests the best two models on eight datasets composed of the whole year dataset, four semester-based datasets, and three season-based datasets. In this work, the semester-based dataset gives the most accurate results with a Root Mean Square Error (RMSE) of 2.51 and the highest coefficient of determination (R 2 ) of 0.48.
- 1 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1754-1760
Occupancy Detection based on WI-FI SysLog Data
Residential and commercial buildings consume almost 27% of the total energy in the United States. Therefore, accurate occupancy detection is necessary and valuable to increase the efficiency of the building’s Heating, Ventilation, and Air Conditioning (HVAC) system while maintaining the occupants’ comfort. The occupancy detection process involves deploying two fundamental types of sensors - environmental sensors (humidity, carbon dioxide, temperature, and light sensors) and specialized sensors (Passive infrared, Radio Frequency Identification sensors, cameras, and WI-FI). These sensors are used with three primary categories of algorithms – analytical, knowledge-based, and data-driven. Our paper uses Syslog data from a university building network to generate eight different datasets after removing non-essential data and retaining only those needed. A framework is then proposed, which consists of two stages. The first stage compares the results of four machine learning algorithms, including K-Nearest Neighbors (KNN), Random Forest (RF), eXtreme Gradient Boosting (XGB), and Logistic Regression (LR), to choose the best two models. In this work, XGB and RF produce the most accurate results. The second stage tests the best two models on eight datasets composed of the whole year dataset, four semester-based datasets, and three season-based datasets. In this work, the semester-based dataset gives the most accurate results with a Root Mean Square Error (RMSE) of 2.51 and the highest coefficient of determination (R 2 ) of 0.48.
- 23 June 2022
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 105 (3) , 1-14
Tuning of Robot Navigation Performance Using Factorial Design
- 23 June 2022
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 105 (3) , 1-14
Tuning of Robot Navigation Performance Using Factorial Design