Pub Date : 2018-10-01DOI: 10.1109/CEIT.2018.8751788
A. Demir, Volkan Sezer
In this study, we present a unified motion planner with low- level controller for continuous control of a differential drive mobile robot. Deep reinforcement agent takes 10 dimensional state vector as input and calculates each wheel’s torque value as a 2 dimensional output vector. These torque values are fed into the dynamic model of the robot, and lastly steering commands are gathered. In previous studies, navigation problem solutions that uses deep - RL methods, have not been considered with agent’s own dynamic constraints, but it has been done by only considering kinematic models. This is not reliable enough for real-world scenarios. In this paper, deep-RL based motion planning is performed by considering both kinematic and dynamic constraints. According to the simulations in a dynamic environment, the agent succesfully navigates through the intersection with 99.6% success rate.
{"title":"Intersection Navigation Under Dynamic Constraints Using Deep Reinforcement Learning","authors":"A. Demir, Volkan Sezer","doi":"10.1109/CEIT.2018.8751788","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751788","url":null,"abstract":"In this study, we present a unified motion planner with low- level controller for continuous control of a differential drive mobile robot. Deep reinforcement agent takes 10 dimensional state vector as input and calculates each wheel’s torque value as a 2 dimensional output vector. These torque values are fed into the dynamic model of the robot, and lastly steering commands are gathered. In previous studies, navigation problem solutions that uses deep - RL methods, have not been considered with agent’s own dynamic constraints, but it has been done by only considering kinematic models. This is not reliable enough for real-world scenarios. In this paper, deep-RL based motion planning is performed by considering both kinematic and dynamic constraints. According to the simulations in a dynamic environment, the agent succesfully navigates through the intersection with 99.6% success rate.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"83 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114042497","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751898
Ayşegül Sarı, Mertcan Cibooglu
Traffic signs play an important role to regulate daily traffic by providing necessary information to the drivers. For unmanned driving systems, real time and robust detection and recognition of traffic signs is one of the main concerns. Therefore, a traffic sign detection and recognition system for autonomous radio controlled cars is proposed. In this work, traditional image processing methods and deep neural networks techniques are combined. First, the online video is streamed from the car camera and the input frame region of interest is detected. Secondly, a convolutional neural network is used to recognize these candidate images. Experimental results show that the proposed system works efficiently up to %87.36 of images. However, calibration is needed for image processing techniques for various environments.
{"title":"Traffic Sign Detection and Recognition System for Autonomous RC Cars","authors":"Ayşegül Sarı, Mertcan Cibooglu","doi":"10.1109/CEIT.2018.8751898","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751898","url":null,"abstract":"Traffic signs play an important role to regulate daily traffic by providing necessary information to the drivers. For unmanned driving systems, real time and robust detection and recognition of traffic signs is one of the main concerns. Therefore, a traffic sign detection and recognition system for autonomous radio controlled cars is proposed. In this work, traditional image processing methods and deep neural networks techniques are combined. First, the online video is streamed from the car camera and the input frame region of interest is detected. Secondly, a convolutional neural network is used to recognize these candidate images. Experimental results show that the proposed system works efficiently up to %87.36 of images. However, calibration is needed for image processing techniques for various environments.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123452893","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751821
Ayberk Cansever, Umut Özel, Önsel Akın, Ali Özmez, Fahrettin Fırat Gönen, Goksen Altuntas Sen, Serdar Kalaycı, Utku Çolak
Today, informing users, customers, members through e-mail, text message or through any other mean is crucial for companies. Information methods get more critical if user/message volume gets higher and message delivery must be on time. Because of performance reasons, the designed systems must be high-available and distributed. In this paper, we discuss how we built a high-performance mailing system by using Apache Kafka as an internal sub-system messaging and how we replaced the existing central database-oriented queuing architecture with this new system. Also, we discuss how we overcame the performance problems of the previous design with new queueing mechanism. About 40 million daily mails are passed through the pipeline successfully after we built the new architecture.
{"title":"A Distributed Message Queuing Mechanism for a Mailing System with High Performance and High Availability","authors":"Ayberk Cansever, Umut Özel, Önsel Akın, Ali Özmez, Fahrettin Fırat Gönen, Goksen Altuntas Sen, Serdar Kalaycı, Utku Çolak","doi":"10.1109/CEIT.2018.8751821","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751821","url":null,"abstract":"Today, informing users, customers, members through e-mail, text message or through any other mean is crucial for companies. Information methods get more critical if user/message volume gets higher and message delivery must be on time. Because of performance reasons, the designed systems must be high-available and distributed. In this paper, we discuss how we built a high-performance mailing system by using Apache Kafka as an internal sub-system messaging and how we replaced the existing central database-oriented queuing architecture with this new system. Also, we discuss how we overcame the performance problems of the previous design with new queueing mechanism. About 40 million daily mails are passed through the pipeline successfully after we built the new architecture.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123589195","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751877
E. Haydar, U. E. Ayten, N. Kahraman
In this paper, a smart sensor system design for home appliances, such as refrigerator, is introduced. The proposed smart sensor system is designed to collect inner temperature and relative humidity level of the appliance along with condition of whether appliance’s door is open or closed. A current sensor is also used to measure the current drawn by the appliance. To control appliance’s connection to the grid a relay is attached to the system. Afterwards all sensor data is sent to a platform on which user can observe sensor values and control relay.
{"title":"Smart Sensor for Home Appliances","authors":"E. Haydar, U. E. Ayten, N. Kahraman","doi":"10.1109/CEIT.2018.8751877","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751877","url":null,"abstract":"In this paper, a smart sensor system design for home appliances, such as refrigerator, is introduced. The proposed smart sensor system is designed to collect inner temperature and relative humidity level of the appliance along with condition of whether appliance’s door is open or closed. A current sensor is also used to measure the current drawn by the appliance. To control appliance’s connection to the grid a relay is attached to the system. Afterwards all sensor data is sent to a platform on which user can observe sensor values and control relay.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125495709","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751872
Ahmetcan Erdogan, Emre Kaplan, A. Leitner, Markus Nager
Validation of highly automated or autonomous vehicles is still a major challenge. One main reason is the need to identify the uncountable potential situations in which the vehicle should be evaluated. Even when they are completely identified, the full-coverage of all these situations is not possible in real-world tests. Virtual validation however can be used to generate infinitesimally changing cases for testing. Nevertheless, the main challenge then becomes the collection and description of realistic test scenarios. This paper proposes an end-to-end approach for scenario generation from various sources. A flexible database structure enables fast and efficient querying that is used to generate an extensive set of test cases. The proposed schema is evaluated end-to-end—from scenario to test case generation to automated simulation and data gathering—populated by a data source based on the experience gained through real world tests. The proposed approach is a further step towards more efficient testing of autonomous functions.
{"title":"Parametrized End-to-End Scenario Generation Architecture for Autonomous Vehicles","authors":"Ahmetcan Erdogan, Emre Kaplan, A. Leitner, Markus Nager","doi":"10.1109/CEIT.2018.8751872","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751872","url":null,"abstract":"Validation of highly automated or autonomous vehicles is still a major challenge. One main reason is the need to identify the uncountable potential situations in which the vehicle should be evaluated. Even when they are completely identified, the full-coverage of all these situations is not possible in real-world tests. Virtual validation however can be used to generate infinitesimally changing cases for testing. Nevertheless, the main challenge then becomes the collection and description of realistic test scenarios. This paper proposes an end-to-end approach for scenario generation from various sources. A flexible database structure enables fast and efficient querying that is used to generate an extensive set of test cases. The proposed schema is evaluated end-to-end—from scenario to test case generation to automated simulation and data gathering—populated by a data source based on the experience gained through real world tests. The proposed approach is a further step towards more efficient testing of autonomous functions.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125986033","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751926
M. Ayaz, M. Tezcan, E. Meşe, K. Yilmaz
Permanent Magnet Synchronous Machines (PMSMs) are getting popular in the automotive applications which are necessary for high power density in low volumes. Also they have a huge usage potential in other industrial applications like robotics, naval applications, space applications etc. In this paper, design parameters of a special PMSM are analyzed which has two independent winding set for motor and generator operations in a simple stator instead of two separate electrical machine. Important design parameters like skew, airgap length, slot-pole combination, magnet height-width which affect the output performance are investigated for multi tasked PMSM. Performance output values are analyzed for two operation situation of multitasked PMSM and optimum design values are determined. These values are obtained from Maxwell-Simplorer softwares where co-simulation analysis has been run.
{"title":"Parametric Optimization of Permanent Magnet Synchronous Machines","authors":"M. Ayaz, M. Tezcan, E. Meşe, K. Yilmaz","doi":"10.1109/CEIT.2018.8751926","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751926","url":null,"abstract":"Permanent Magnet Synchronous Machines (PMSMs) are getting popular in the automotive applications which are necessary for high power density in low volumes. Also they have a huge usage potential in other industrial applications like robotics, naval applications, space applications etc. In this paper, design parameters of a special PMSM are analyzed which has two independent winding set for motor and generator operations in a simple stator instead of two separate electrical machine. Important design parameters like skew, airgap length, slot-pole combination, magnet height-width which affect the output performance are investigated for multi tasked PMSM. Performance output values are analyzed for two operation situation of multitasked PMSM and optimum design values are determined. These values are obtained from Maxwell-Simplorer softwares where co-simulation analysis has been run.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130361849","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751914
H. Mekki, O. Benzineb, L. Chrifi-Alaoui, D. Boukhetala, M. Tadjine
This paper proposes a new robust fault reconstruction and estimation design for a class of induction motor (IM) drives. In this purpose, a nonlinear controller based new Integral backstepping strategy is adopted to deal with speed and flux tracking problems under un-certainties (parameter variations and external load disturbance). Later, will be combined with a modified sliding mode observer (SMO) based online faults reconstruction, employed for simultaneously state and actuator faults estimations. The closed-loop stability is guarantee, using the Lyapunov stability theory. Simulation results are given to illustrate the proposed approach performances.
{"title":"Robust Fault Reconstruction with Integral Backstepping Control Design for IM Drives","authors":"H. Mekki, O. Benzineb, L. Chrifi-Alaoui, D. Boukhetala, M. Tadjine","doi":"10.1109/CEIT.2018.8751914","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751914","url":null,"abstract":"This paper proposes a new robust fault reconstruction and estimation design for a class of induction motor (IM) drives. In this purpose, a nonlinear controller based new Integral backstepping strategy is adopted to deal with speed and flux tracking problems under un-certainties (parameter variations and external load disturbance). Later, will be combined with a modified sliding mode observer (SMO) based online faults reconstruction, employed for simultaneously state and actuator faults estimations. The closed-loop stability is guarantee, using the Lyapunov stability theory. Simulation results are given to illustrate the proposed approach performances.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127199489","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751916
Houssem E M Triki, M. Lagoun, E. Bdirina
This paper has for objective, the maximization of a one-body wave energy converter using model predictive control. Where two cases are tested. In the second case, a proposition of using the observer model to extract the control law in order to ameliorate control characteristics and reduce the costs of using sensors. The suggested approach is applied to our system and compared to the first case, which represent the classical approach, through simulation results.
{"title":"Wave energy converter control using observed model predictive control","authors":"Houssem E M Triki, M. Lagoun, E. Bdirina","doi":"10.1109/CEIT.2018.8751916","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751916","url":null,"abstract":"This paper has for objective, the maximization of a one-body wave energy converter using model predictive control. Where two cases are tested. In the second case, a proposition of using the observer model to extract the control law in order to ameliorate control characteristics and reduce the costs of using sensors. The suggested approach is applied to our system and compared to the first case, which represent the classical approach, through simulation results.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129504962","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751847
M. Derakhshan, M. Haeri
In this paper a method is presented to estimate the leak site in sloped transmission lines using a closed solution of the transmission line equations and linear observer. First, the equations are solved for a constant input state, and then the initial estimation of the leak site is presented for variable input state and the leak and pressure coefficient in the leak site is determined with the help of equations obtained from the closed solution using the estimation of the leak site. Then, the nonlinear model is determined by discretizing the transmission line equations with respect to the site variable considering a floating point for the leak site in the observer, designing a linear observer using the initial estimation obtained from the closed solution, and linearizing the observers equations at the obtained point.
{"title":"Detection of Leak Site in Sloped Transmission Lines","authors":"M. Derakhshan, M. Haeri","doi":"10.1109/CEIT.2018.8751847","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751847","url":null,"abstract":"In this paper a method is presented to estimate the leak site in sloped transmission lines using a closed solution of the transmission line equations and linear observer. First, the equations are solved for a constant input state, and then the initial estimation of the leak site is presented for variable input state and the leak and pressure coefficient in the leak site is determined with the help of equations obtained from the closed solution using the estimation of the leak site. Then, the nonlinear model is determined by discretizing the transmission line equations with respect to the site variable considering a floating point for the leak site in the observer, designing a linear observer using the initial estimation obtained from the closed solution, and linearizing the observers equations at the obtained point.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126384850","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 : 2018-10-01DOI: 10.1109/CEIT.2018.8751765
S. Ozer, Senem Yildiz, A. Iftar
Smith predictor-based and optimization-based controller design for a DC-motor control system with a time-delay in the feedback loop is considered. It is shown that, compared to the Smith predictor-based controller, the optimization-based controller can preserve stability against larger variations in the system parameters and the time-delay. Furthermore, by simulations and actual experiments, it is verified that the optimization-based controller outperforms the Smith predictor-based controller when the system parameters and, especially, the time-delay varies.
{"title":"Performance Analysis of a DC-Motor Control System with Time-Delay: Smith Predictor vs Optimization-based Controller Design","authors":"S. Ozer, Senem Yildiz, A. Iftar","doi":"10.1109/CEIT.2018.8751765","DOIUrl":"https://doi.org/10.1109/CEIT.2018.8751765","url":null,"abstract":"Smith predictor-based and optimization-based controller design for a DC-motor control system with a time-delay in the feedback loop is considered. It is shown that, compared to the Smith predictor-based controller, the optimization-based controller can preserve stability against larger variations in the system parameters and the time-delay. Furthermore, by simulations and actual experiments, it is verified that the optimization-based controller outperforms the Smith predictor-based controller when the system parameters and, especially, the time-delay varies.","PeriodicalId":357613,"journal":{"name":"2018 6th International Conference on Control Engineering & Information Technology (CEIT)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124464964","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：