Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156726
Tomás Soucek, A. Richter, J. Morava, L. Slavík
The paper introduces a physical principles to serve as a basis of the following clinical study, which goal is to analyze an interaction between the specific patient with implanted pacemaker and possibly dangerous source of interference electromagnetic fields (EMF). The patient works as an operator of industry machine tool - surface grinder. Manufacturers of cardiac implantable electronic devices (CIED) consider industrial equipment of this type risky for patients and they recommend to avoid it. In this paper we present a mapping of patient's workspace to determine potentially dangerous sources of EMF and assess their relevance in context of possible risk for CIED and related legislation and limits. This paper should describe the case and serve as a first step in the clinical study.
{"title":"The mapping of electromagnetic field of machine tool for assessment of its influence on the pacemaker","authors":"Tomás Soucek, A. Richter, J. Morava, L. Slavík","doi":"10.1109/HORA58378.2023.10156726","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156726","url":null,"abstract":"The paper introduces a physical principles to serve as a basis of the following clinical study, which goal is to analyze an interaction between the specific patient with implanted pacemaker and possibly dangerous source of interference electromagnetic fields (EMF). The patient works as an operator of industry machine tool - surface grinder. Manufacturers of cardiac implantable electronic devices (CIED) consider industrial equipment of this type risky for patients and they recommend to avoid it. In this paper we present a mapping of patient's workspace to determine potentially dangerous sources of EMF and assess their relevance in context of possible risk for CIED and related legislation and limits. This paper should describe the case and serve as a first step in the clinical study.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130734714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156799
Sejma Cicak, Umut Avci
Imbalanced data is a common problem in many areas, and it can have significant impacts on the performance and generalizability of machine learning models. This is because the models fail to create a good representation of the examples in the minority class. This study aims at improving the classification success for the predictive maintenance tasks in which the data is generally imbalanced. To this end, we use resampling methods that target creating balanced data. We present various oversampling and undersampling techniques and apply them to both synthetic and real-world datasets. We then perform classification experiments with imbalanced and balanced datasets by using different classifiers. The performances of different classifiers have been compared. More importantly, we evaluate the effectiveness of resampling techniques to provide insights into their usefulness in handling class imbalance. Our study contributes to the growing body of literature on addressing the class imbalance in classification tasks and provides practical guidance for selecting appropriate sampling methods based on the characteristics of the dataset.
{"title":"Handling Imbalanced Data in Predictive Maintenance: A Resampling-Based Approach","authors":"Sejma Cicak, Umut Avci","doi":"10.1109/HORA58378.2023.10156799","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156799","url":null,"abstract":"Imbalanced data is a common problem in many areas, and it can have significant impacts on the performance and generalizability of machine learning models. This is because the models fail to create a good representation of the examples in the minority class. This study aims at improving the classification success for the predictive maintenance tasks in which the data is generally imbalanced. To this end, we use resampling methods that target creating balanced data. We present various oversampling and undersampling techniques and apply them to both synthetic and real-world datasets. We then perform classification experiments with imbalanced and balanced datasets by using different classifiers. The performances of different classifiers have been compared. More importantly, we evaluate the effectiveness of resampling techniques to provide insights into their usefulness in handling class imbalance. Our study contributes to the growing body of literature on addressing the class imbalance in classification tasks and provides practical guidance for selecting appropriate sampling methods based on the characteristics of the dataset.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132225899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156682
Mehmet Batuhan Özdaş, Fatih Uysal, F. Hardalaç
Automatic object detection is important in the military industry. Since these objects are small and camouflaged, that is, they are not clear, it becomes even more important that they appear clear and large. Therefore, in order to facilitate object detection algorithms in the field of the military industry, we present a model that obtains high-resolution and high-dimensional images from low-resolution and low-dimensional images. The presented model is a combination of fast super-resolution convolutional neural networks and the VGG16 model, which is widely used in the literature. Due to the limited data in the field of the military industry, the dataset was collected manually from the internet. Our dataset, which has 900 images in total, has been reproduced with certain data augmentation techniques. For model training, low-dimensional images were obtained from the collected high-dimensional images by the bicubic interpolation method. After model training, a BRISQUE score of 47.81 was obtained.
{"title":"Super Resolution Image Acquisition for Object Detection in the Military Industry","authors":"Mehmet Batuhan Özdaş, Fatih Uysal, F. Hardalaç","doi":"10.1109/HORA58378.2023.10156682","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156682","url":null,"abstract":"Automatic object detection is important in the military industry. Since these objects are small and camouflaged, that is, they are not clear, it becomes even more important that they appear clear and large. Therefore, in order to facilitate object detection algorithms in the field of the military industry, we present a model that obtains high-resolution and high-dimensional images from low-resolution and low-dimensional images. The presented model is a combination of fast super-resolution convolutional neural networks and the VGG16 model, which is widely used in the literature. Due to the limited data in the field of the military industry, the dataset was collected manually from the internet. Our dataset, which has 900 images in total, has been reproduced with certain data augmentation techniques. For model training, low-dimensional images were obtained from the collected high-dimensional images by the bicubic interpolation method. After model training, a BRISQUE score of 47.81 was obtained.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"301 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134145794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156714
Ahmet Seyfullah Güneş, Saffet Vatansever
Electricity network frequency (ENF) is the frequency of electrical voltage transmitted by power distribution lines, with a nominal value of 50 Hz in most of the world and a nominal value of 60 Hz in the vast majority of America. The ENF makes continuous oscillations within certain limits around the nominal value depending on the supply-demand power imbalance in the network. These time-dependent changes in ENF are called the ENF signal. Although the ENF signal may show similarities in short time intervals, it becomes unique in large time intervals. The ENF signal is intrinsically integrated into audio and video recordings under certain conditions. The fact that the ENF signal shows different characteristics in different networks and is unique depending on time allows researchers to make inferences about the file content integrity, together with the location and time information of the audio and video files. In this study, it is discussed how to detect modifications in the file content and metadata using ENF within the scope of ENF-based forensic analysis of audio and video. In this context, existing ENF applications in the literature and the potential ENF usage areas are examined and analyzed.
{"title":"A Review of Electrical Network Frequency (ENF) Based Applications in Media Forensics","authors":"Ahmet Seyfullah Güneş, Saffet Vatansever","doi":"10.1109/HORA58378.2023.10156714","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156714","url":null,"abstract":"Electricity network frequency (ENF) is the frequency of electrical voltage transmitted by power distribution lines, with a nominal value of 50 Hz in most of the world and a nominal value of 60 Hz in the vast majority of America. The ENF makes continuous oscillations within certain limits around the nominal value depending on the supply-demand power imbalance in the network. These time-dependent changes in ENF are called the ENF signal. Although the ENF signal may show similarities in short time intervals, it becomes unique in large time intervals. The ENF signal is intrinsically integrated into audio and video recordings under certain conditions. The fact that the ENF signal shows different characteristics in different networks and is unique depending on time allows researchers to make inferences about the file content integrity, together with the location and time information of the audio and video files. In this study, it is discussed how to detect modifications in the file content and metadata using ENF within the scope of ENF-based forensic analysis of audio and video. In this context, existing ENF applications in the literature and the potential ENF usage areas are examined and analyzed.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132154657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156670
Thulfiqar Muayad Hameedi, Gulsum Akkuzu Kaya
Images are one of the most widely used multimedia in the correspondence between people, as some of the characteristics of these images can be used to hide important messages. Each image has different characteristics, and the method of concealment changes depending on the characteristics of the image used. In this research, an algorithm was proposed to increase the efficiency of the data embedding algorithm by relying on some of the characteristics of the colored digital image. First, the color image is dismantled to the basic color layers (red, green, blue). Then, the amount of variation in each layer is measured by using image processing techniques. After that, the high contrast layer is identified and used as a cover to include the message to be included, while the other two layers are used as a key to the encryption algorithm that is applied to the text before the embedding process to increase data security.The method of concealment depends on the first and second bit values in the selected layer as a cover for the embedding process. Three criteria were used to measure the efficiency of the proposed algorithm.
{"title":"Enhanced Data Hiding Using Some Attribute of Color Image","authors":"Thulfiqar Muayad Hameedi, Gulsum Akkuzu Kaya","doi":"10.1109/HORA58378.2023.10156670","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156670","url":null,"abstract":"Images are one of the most widely used multimedia in the correspondence between people, as some of the characteristics of these images can be used to hide important messages. Each image has different characteristics, and the method of concealment changes depending on the characteristics of the image used. In this research, an algorithm was proposed to increase the efficiency of the data embedding algorithm by relying on some of the characteristics of the colored digital image. First, the color image is dismantled to the basic color layers (red, green, blue). Then, the amount of variation in each layer is measured by using image processing techniques. After that, the high contrast layer is identified and used as a cover to include the message to be included, while the other two layers are used as a key to the encryption algorithm that is applied to the text before the embedding process to increase data security.The method of concealment depends on the first and second bit values in the selected layer as a cover for the embedding process. Three criteria were used to measure the efficiency of the proposed algorithm.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"176 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133215731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156657
Marwa Khanom Nurtaj, Rafsan Bari Shafin, M. Hasan, Krittika Roy, M. S. Hossain Khan, Rashedul Amin Tuhin, Md. Mohsin Uddin
In this paper, we investigate the efficacy of various cutting-edge models for update summarization on the TAC 2009 dataset. To construct abstractive and extractive summaries of news items, we use the T5 Transformer model and Textrank + Pegasus model. Our goal is to assess how well these models capture key information from updates and generate coherent and useful summaries. Here we use conventional assessment measures such as ROUGE to assess the performance of the models. We analyze the fluency, coherence, and informativeness of generated summaries from the T5 Transformer model, Textrank + Pegasus, and TensorFlow models against human-authored gold summaries.
{"title":"Enhancing Performance of Abstractive Multi-Document Update Summarization on TAC Dataset","authors":"Marwa Khanom Nurtaj, Rafsan Bari Shafin, M. Hasan, Krittika Roy, M. S. Hossain Khan, Rashedul Amin Tuhin, Md. Mohsin Uddin","doi":"10.1109/HORA58378.2023.10156657","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156657","url":null,"abstract":"In this paper, we investigate the efficacy of various cutting-edge models for update summarization on the TAC 2009 dataset. To construct abstractive and extractive summaries of news items, we use the T5 Transformer model and Textrank + Pegasus model. Our goal is to assess how well these models capture key information from updates and generate coherent and useful summaries. Here we use conventional assessment measures such as ROUGE to assess the performance of the models. We analyze the fluency, coherence, and informativeness of generated summaries from the T5 Transformer model, Textrank + Pegasus, and TensorFlow models against human-authored gold summaries.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132618394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156673
Tanvir Rahman, Miah Mohammad Asif Syeed, Maisha Farzana, Ishadie Namir, Ipshita Ishrar, Meherin Hossain Nushra, Bhoktear Mahbub Khan
India experiences recurrent natural disasters in the form of floods, which result in substantial destruction of both human life and property. Accurately predicting the onset and progression of floods in real-time is crucial for minimizing their impact. This research paper focuses on a comparative study of various machine learning models for flood prediction in India. The evaluated models include K-Nearest Neighbor (KNN), Support Vector Classifier (SVC), Decision tree Classifier, Binary Logistic Regression, and Stacked Generalization (Stacking). We used a dataset of rainfall to train and test the models. Our results indicate that the stacked generalization model outperforms the other models, achieving an accuracy of 93.3% and Standard Deviation of 0.098. Our findings suggest that machine learning models can provide accurate and timely flood predictions, enabling disaster management authorities to take appropriate measures to minimize damage and save lives.
{"title":"Flood Prediction Using Ensemble Machine Learning Model","authors":"Tanvir Rahman, Miah Mohammad Asif Syeed, Maisha Farzana, Ishadie Namir, Ipshita Ishrar, Meherin Hossain Nushra, Bhoktear Mahbub Khan","doi":"10.1109/HORA58378.2023.10156673","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156673","url":null,"abstract":"India experiences recurrent natural disasters in the form of floods, which result in substantial destruction of both human life and property. Accurately predicting the onset and progression of floods in real-time is crucial for minimizing their impact. This research paper focuses on a comparative study of various machine learning models for flood prediction in India. The evaluated models include K-Nearest Neighbor (KNN), Support Vector Classifier (SVC), Decision tree Classifier, Binary Logistic Regression, and Stacked Generalization (Stacking). We used a dataset of rainfall to train and test the models. Our results indicate that the stacked generalization model outperforms the other models, achieving an accuracy of 93.3% and Standard Deviation of 0.098. Our findings suggest that machine learning models can provide accurate and timely flood predictions, enabling disaster management authorities to take appropriate measures to minimize damage and save lives.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"260 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115598369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This research paper proposes a comprehensive approach to enhance the well-being of dogs through a range of innovative technologies. Firstly, we develop an automated system for dog breed and age identification using a Convolutional Neural Network (CNN) and a transfer learning model. This system aims to provide an efficient and reliable solution for dog owners and new adopters who are interested in discovering more about their canine companions. Secondly, we propose the development of a system that uses Reinforcement Learning to generate personalized meal plans based on a variety of factors such as the dog's breed, age, weight, health status, and emotional state. The system aims to provide dog owners with a reliable and effective tool for generating personalized meal plans that will enhance their pets' overall health and well-being. Thirdly, we present a dog disease recognition application that utilizes an artificial neural network (ANN) for identifying dog diseases based on their symptoms. Lastly, we introduce a real-time remote dog monitoring system using loT devices with edge computing to detect aggressive and anxious sounds. Our system provides an accurate classification of dog sounds related to aggression and anxiety, which can help dog owners detect and respond to potential issues early on. This research aims to provide dog owners and veterinarians with a range of technologies that can help them better understand and care for their furry friends.
{"title":"Advancing Canine Health and Care: A Multifaceted Approach using Machine Learning","authors":"Yasith Wimukthi, Hashen Kottegoda, Dilshan Andaraweera, Pabasara Palihena, H.S.M.H. Fernando, Darshana Kasthurirathnae","doi":"10.1109/HORA58378.2023.10155781","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10155781","url":null,"abstract":"This research paper proposes a comprehensive approach to enhance the well-being of dogs through a range of innovative technologies. Firstly, we develop an automated system for dog breed and age identification using a Convolutional Neural Network (CNN) and a transfer learning model. This system aims to provide an efficient and reliable solution for dog owners and new adopters who are interested in discovering more about their canine companions. Secondly, we propose the development of a system that uses Reinforcement Learning to generate personalized meal plans based on a variety of factors such as the dog's breed, age, weight, health status, and emotional state. The system aims to provide dog owners with a reliable and effective tool for generating personalized meal plans that will enhance their pets' overall health and well-being. Thirdly, we present a dog disease recognition application that utilizes an artificial neural network (ANN) for identifying dog diseases based on their symptoms. Lastly, we introduce a real-time remote dog monitoring system using loT devices with edge computing to detect aggressive and anxious sounds. Our system provides an accurate classification of dog sounds related to aggression and anxiety, which can help dog owners detect and respond to potential issues early on. This research aims to provide dog owners and veterinarians with a range of technologies that can help them better understand and care for their furry friends.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115618723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156725
K. Shoilekova, Boyana Ivanova
This article describes the logic model of an application for moving from one point to another using different types of transport. The presented models aim to describe the structure of the application, and some of them are used to describe the behavior of the system. An important point in the modeling and implementation of the application is the accurate mapping of the stops of the various types of public transport and the timetable of their transport lines. The presence of a GPS receiver in vehicles allows for dynamic changes in travel time, which in turn can influence the choice of option/route for moving from one point to another.
{"title":"Modeling a system determining the fastest way to get from one point to another by public transport","authors":"K. Shoilekova, Boyana Ivanova","doi":"10.1109/HORA58378.2023.10156725","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156725","url":null,"abstract":"This article describes the logic model of an application for moving from one point to another using different types of transport. The presented models aim to describe the structure of the application, and some of them are used to describe the behavior of the system. An important point in the modeling and implementation of the application is the accurate mapping of the stops of the various types of public transport and the timetable of their transport lines. The presence of a GPS receiver in vehicles allows for dynamic changes in travel time, which in turn can influence the choice of option/route for moving from one point to another.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114057172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-08DOI: 10.1109/HORA58378.2023.10156788
Utkarsh Gupta, Anish Gorantiwar, S. Taheri
Vehicle suspension systems are crucial in optimizing the vehicle's ride comfort and road holding properties. Semi-active and active suspension systems play a significant role in bridging the gap in achieving the desired vertical dynamic characteristics of the vehicles compared to the traditional non-controllable and controllable suspension systems. Conventional controllable suspension systems utilize either a completely data-driven approach toward developing a control function or a classical control framework that enables the variation of the damping characteristics of the suspension system. These approaches suffer from the volatile nature of the driving conditions due to variations in speed, tire load, road surface, road grade, banking angles, etc. In this paper, a novel approach toward the control of the vertical dynamic characteristics of a vehicle has been proposed based on a fusion of theoretical knowledge with experimental data in a Physics-guided Machine Learning setting. A proposed three-system architecture comprised a model-based estimation, actual data-driven model training, and experimental validation. The proposed Physics-guided architecture has been implemented using simulated data and validated using experimental data from a Shock Dyno Suspension test rig. The developed algorithm draws its roots from a base-excitation suspension model and feeds upon the sprung and unsprung mass accelerations to control the damping characteristics of a semi-active suspension system in real-time. This control framework has been compared with the classical suspension control algorithms - Skyhook and Groundhook control based on the performance metrics of comfort cost about the chassis frequency zone.
{"title":"Vehicle Suspension Control using Physics Guided Machine Learning","authors":"Utkarsh Gupta, Anish Gorantiwar, S. Taheri","doi":"10.1109/HORA58378.2023.10156788","DOIUrl":"https://doi.org/10.1109/HORA58378.2023.10156788","url":null,"abstract":"Vehicle suspension systems are crucial in optimizing the vehicle's ride comfort and road holding properties. Semi-active and active suspension systems play a significant role in bridging the gap in achieving the desired vertical dynamic characteristics of the vehicles compared to the traditional non-controllable and controllable suspension systems. Conventional controllable suspension systems utilize either a completely data-driven approach toward developing a control function or a classical control framework that enables the variation of the damping characteristics of the suspension system. These approaches suffer from the volatile nature of the driving conditions due to variations in speed, tire load, road surface, road grade, banking angles, etc. In this paper, a novel approach toward the control of the vertical dynamic characteristics of a vehicle has been proposed based on a fusion of theoretical knowledge with experimental data in a Physics-guided Machine Learning setting. A proposed three-system architecture comprised a model-based estimation, actual data-driven model training, and experimental validation. The proposed Physics-guided architecture has been implemented using simulated data and validated using experimental data from a Shock Dyno Suspension test rig. The developed algorithm draws its roots from a base-excitation suspension model and feeds upon the sprung and unsprung mass accelerations to control the damping characteristics of a semi-active suspension system in real-time. This control framework has been compared with the classical suspension control algorithms - Skyhook and Groundhook control based on the performance metrics of comfort cost about the chassis frequency zone.","PeriodicalId":247679,"journal":{"name":"2023 5th International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123727863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：