Saber Kazeminasab

Saleh Afroogh, Seyd Ishtiaque Ahmed, Petra Ahrweiler et al. · cmu

This study provides a cross-disciplinary examination of Explainable Artificial Intelligence (XAI) approaches-focusing on deep neural networks (DNNs) and large language models (LLMs)-and identifies empirical and conceptual limitations in current XAI. We discuss critical symptoms that stem from deeper root causes (i.e., two paradoxes, two conceptual confusions, and five false assumptions). These fundamental problems within the current XAI research field reveal three insights: experimentally, XAI exhibits significant flaws; conceptually, it is paradoxical; and pragmatically, further attempts to reform the paradoxical XAI might exacerbate its confusion-demanding fundamental shifts and new research directions. To move beyond XAI's limitations, we propose a four-pronged synthesized paradigm shift toward reliable and certified AI development. These four components include: verification-focused Interactive AI (IAI) to establish scientific community protocols for certifying AI system performance rather than attempting post-hoc explanations, AI Epistemology for rigorous scientific foundations, User-Sensible AI to create context-aware systems tailored to specific user communities, and Model-Centered Interpretability for faithful technical analysis-together offering comprehensive post-XAI research directions.

Moein Razavi, Samira Ziyadidegan, Reza Jahromi et al.

Background: Mental stress and its consequent mental disorders (MDs) are significant public health issues. With the advent of machine learning (ML), there's potential to harness computational techniques for better understanding and addressing these problems. This review seeks to elucidate the current ML methodologies employed in this domain to enhance the detection, prediction, and analysis of mental stress and MDs. Objective: This review aims to investigate the scope of ML methodologies used in the detection, prediction, and analysis of mental stress and MDs. Methods: Utilizing a rigorous scoping review process with PRISMA-ScR guidelines, this investigation delves into the latest ML algorithms, preprocessing techniques, and data types used in the context of stress and stress-related MDs. Results and Discussion: A total of 98 peer-reviewed publications were examined. The findings highlight that Support Vector Machine (SVM), Neural Network (NN), and Random Forest (RF) models consistently exhibit superior accuracy and robustness among ML algorithms. Physiological parameters such as heart rate measurements and skin response are prevalently used as stress predictors due to their rich explanatory information and ease of data acquisition. Dimensionality reduction techniques, including mappings, feature selection, filtering, and noise reduction, are frequently observed as crucial steps preceding the training of ML algorithms. Conclusion: This review identifies significant research gaps and outlines future directions for the field. These include model interpretability, model personalization, the incorporation of naturalistic settings, and real-time processing capabilities for the detection and prediction of stress and stress-related MDs. Keywords: Machine Learning; Deep Learning; Data Preprocessing; Stress Detection; Stress Prediction; Stress Monitoring; Mental Disorders

Saber Kazeminasab, M. Katherine Banks

In this paper, we propose an operation procedure for our previously developed in-pipe robotic system that is used for water quality monitoring in water distribution systems (WDS). The proposed operation procedure synchronizes a developed wireless communication system that is suitable for harsh environments of soil, water, and rock with a multi-phase control algorithm. The new wireless control algorithm facilitates smart navigation and near real-time wireless data transmission during operation for our in-pipe robot in WDS. The smart navigation enables the robot to pass through different configurations of the pipeline with long inspection capability with a battery in which is mounted on the robot. To this end, we have divided the operation procedure into five steps that assign a specific motion control phase and wireless communication task to the robot. We describe each step and the algorithm associated with that step in this paper. The proposed robotic system defines the configuration type in each pipeline with the pre-programmed pipeline map that is given to the robot before the operation and the wireless communication system. The wireless communication system includes some relay nodes that perform bi-directional communication in the operation procedure. The developed wireless robotic system along with operation procedure facilitates localization and navigation for the robot toward long-distance inspection in WDS.

Saber Kazeminasab, M Katherine Banks

Incident in water distribution systems (WDS) cause water loss and water contamination that requires the utility managers to assess the condition of pipelines in a timely manner. However, pipelines are long and access to all parts of it is a chal-lenging task; current in-pipe robots have the limitations of short-distance inspection and inability to operate in-service networks. In this work, we improve the design of our previously developed in-pipe robot and analyze the effect of line pressure and relative velocity on the robot during operation with computational fluid dynamics (CFD) simulations. An extreme scenario for robot operation is defined and we estimate the minimum inspection distance for the robot with one turn of battery charge that is 5400m. A multi-phase motion controller is proposed that ensures reliable motion at straight and non-straight configurations of pipeline and also stabilized configuration with zero velocity at junctions. We also propose a localization and navigation method based on particle filtering and combine it with the proposed multi-phase motion controller. In this method, the map of the operation is provided to the robot and the robot localizes itself based on the map and the particle filtering method. Furthermore, the robot navigates different configurations of pipelines by switching between different phases of the motion controller algorithm that is performed by the particle filter algorithm. The experiment and simulation results show that the robot along with the navigation shows a promising solution towards long-distance inspection of pipelines by in-pipe robots.

Saber Kazeminasab, Vahid Janfaza, Moein Razavi et al.

In-pipe robots are promising solutions for condition assessment, leak detection, water quality monitoring in a variety of other tasks in pipeline networks. Smart navigation is an extremely challenging task for these robots as a result of highly uncertain and disturbing environment for operation. Wireless communication to control these robots during operation is not feasible if the pipe material is metal since the radio signals are destroyed in the pipe environment, and hence, this challenge is still unsolved. In this paper, we introduce a method for smart navigation for our previously designed in-pipe robot [1] based on particle filtering and a two-phase motion controller. The robot is given the map of the operation path with a novel approach and the particle filtering determines the straight and non-straight configurations of the pipeline. In the straight paths, the robot follows a linear quadratic regulator (LQR) and proportional-integral-derivative (PID) based controller that stabilizes the robot and tracks a desired velocity. In non-straight paths, the robot follows the trajectory that a motion trajectory generator block plans for the robot. The proposed method is a promising solution for smart navigation without the need for wireless communication and capable of inspecting long distances in water distribution systems.

Saber Kazeminasab, Ali Akbari, Roozbeh Jafari et al.

Leak detection and water quality monitoring are requirements and challenging tasks in Water Distribution Systems (WDS). In-line robots are designed for this aim. In our previous work, we designed an in-pipe robot [1]. In this research, we present the design of the central processor, characterize and control the robot based on the condition of operation in a highly pressurized environment of pipelines with the presence of high-speed flow. To this aim, an extreme operation condition is simulated with computational fluid dynamics (CFD) and the spring mechanism is characterized to ensure sufficient stabilizing force during operation based on the extreme operating condition. Also, an end-to-end method is suggested for power considerations for our robot that calculates minimum battery capacity and operation duration in the extreme operating condition. Finally, we design a novel LQR-PID based controller based on the system auxiliary matrices that retain the robot stability inside the pipeline against disturbances and uncertainties during operation. The ADAMS-MATLAB co-simulation of the robot-controller shows the rotational velocity with -4 degree/sec and +3 degree/sec margin around x, y, and z axes while the system tracks different desired velocities in pipelines (i.e. 0.12m/s, 0.17m/s, and 0.35m/s). Also, experimental results for four iterations in a 14-inch diameter PVC pipe show that the controller brings initial values of stabilizing states to zero and oscillate around it with a margin of 2 degrees and the system tracks desired velocities of 0.1m/s, 0.2m/s, 0.3m/s, and 0.35m/s in which makes the robot dexterous in uncertain and highly disturbed the environment of pipelines during operation.