Search

The transition to non-fossil fuels brings with its basic challenges in battery technologies. Due to their efficiency, one of the areas where Li-ion batteries are widely used is electric vehicles (E...

With the onset of the COVID-19 epidemic, distance education technologies have become critical for the whole world. With the development of these technologies, it is now possible for us to access educational tools and contents from anywhere, anytime. We see that robotic systems are also increasing in importance in this age of isolation. This situation requires that robotic education be expanded and made easily accessible. In this study, a few examples of mobile robots that can be used for robotic education and the virtual robotics laboratory system Uplat, which will make the use of these robots quite accessible, are described. The system allows users (or students) to access the virtual robotics lab via the web, control the simulated robots through interfaces, analyze and evaluate the experimental results. Compared to traditional learning methods, UPlat is an easier and better learning platform with open-source ROS tutorials. These tutorials can be launched from anywhere without dealing with complex Linux and ROS installations. All tools required for training are presented in a simplified interface, taking user experiences into account. These tools are presented to the user through the web interface using cloud technology and container technology. In this system, task trainings of four different types of mobile robots (EvaMars, EvaSec, ATEKS and AGV-OTA) designed to be used in many different robotic fields are presented with detailed learning outcomes. Thanks to these robots and improved robotics trainings, a wide perspective of robotics usage will be presented to those who will receive training in this field.

As our dependence on automated systems grows, so does the need for guaranteeing their safety, cybersecurity, and privacy (SCP). Dedicated methods for verification and validation (V&V) must be used to this end and it is necessary that the methods and their characteristics can be clearly differentiated. This can be achieved via method classifications. However, we have experienced that existing classifications are not suitable to categorise V&V methods for SCP of automated systems. They do not pay enough attention to the distinguishing characteristics of this system type and of these quality concerns. As a solution, we present a new classification developed in the scope of a large-scale industry-academia project. The classification considers both the method type, e.g., testing, and the concern addressed, e.g., safety. Over 70 people have successfully used the classification on 53 methods. We argue that the classification is a more suitable means to categorise V&V methods for SCP of automated systems and that it can help other researchers and practitioners.

In smart manufacturing industry, health analysis and forecasting of degradation starting point has become an increasingly crucial research area. Prognostics-aware systems for health analysis aim to integrate health information and knowledge about the future operating conditions into the process of selecting subsequent actions for the system. Developments in smart manufacturing as well as deep learning-based prognostics provide new opportunities for health analysis and degradation starting point forecasting. Rotating machines have many critical components like spinning, drilling, rotating, etc. and they need to be forecasted for failure or degradation starting times. Moreover, bearings are the most important sub-components of rotating machines. In this study, a convolutional neural network is used for forecasting of degradation starting point of bearings by experimenting with Nasa Bearing Dataset. Although convolutional neural networks (CNNs) are utilized widely for 2D images, 1-dimensional convolutional filters may also be embedded in processing temporal data, such as time-series. In this work, we developed one such autoencoder network which consists of stacked convolutional layers as a contribution to the community. Besides, in evaluation of test results, L10 bearing life criteria is used for threshold of degradation starting point. Tests are conducted for all bearings and results are shown in different figures. In the test results, proposed method is found to be effective in forecasting bearing degradation starting points.

Verification and validation (V&V) of systems, and system of systems, in an industrial context has never been as important as today. The recent developments in automated cyber-physical systems, digital twin environments, and Industry 4.0 applications require effective and comprehensive V&V mechanisms. Verification and Validation of Automated Systems' Safety and Security (VALU3S), a Horizon 2020 Electronic Components and Systems for European Leadership Joint Undertaking (ECSEL-JU) project started in May 2020, aims to create and evaluate a multi-domain V&V framework that facilitates evaluation of automated systems from component level to system level, with the aim of reducing the time and effort needed to evaluate these systems. VALU3S focuses on V&V for the requirements of safety, cybersecurity, and privacy (SCP). This paper mainly focuses on the elaboration of one of the 13 use cases of VALU3S to identify the SCP issues in an automated robot inspection cell that is being actively used for the quality control assessment of automotive body-in-white. The joint study here embarks on a collaborative approach that puts the V&V methods and workflows for the robotic arms safety trajectory planning and execution, fault injection techniques, cyber-physical security vulnerability assessment, anomaly detection, and SCP countermeasures required for remote control and inspection. The paper also presents cross-links with ECSEL-JU goals and the current advancements in the market and scientific and technological state-of-play.

Recent developments in electric vehicle technology have lead to an increase in the use of electric vehicles. With the increase in the use of electric vehicles, it is necessary to secure the information and money flow in the process from production to consumption. In this study, an energy ecosystem in the Ethereum Blockchain network, which records all processes from the generation of electricity to the end user, is designed. Furthermore, this study includes energy producers, consumers, distributors, dealers, electric charge station and electric vehicle users. Transactions between users are provided using smart contracts. Smart contracts eliminate third-party contacts and all transactions with the decentralized application created are recorded on the Blockchain network. Recorded transactions are maintained in accordance with the principles of confidentiality, integrity and availability through Blockchain and smart contracts. In the application developed with the use of smart contracts, the user can access information such as location, price list, payment type, charging mode, charging type and plug types of the related charging stations. PROMETHEE, which is a multi-criteria decision-making method, is used in case there is more than one offer suitable for a user request. In this method, the weights required for the criteria are determined according to the profile of a user.

Augmented Reality (AR) helps to increase the perception of reality by adding digital data to the real world. Nowadays, AR is used effectively in many areas from entertainment sectors to health and education applications. In this study, a mobile application has been developed by using Augmented Reality technology for maintenance and repair of the mobile robot (evarobot). In this study, battery, motor, sensor and electronic card replacement scenarios of Evarobot and maintenance and repair steps were determined. These steps are supported with visuals in accordance with the instructions given and are intended to enable the user to carry out the maintenance process easily without the need for an expert or maintenance document. At the end of the study, the digitization of the experience and the transfer of the information without getting lost has been realized in an easy way.

WOS: 000441749700004, Indoor positioning system is an active research area. There are various performance metrics such as accuracy, computation time, precision, and f-score in machine learning based indoor positioning systems. The aim of this study is to present a multi-criteria decision strategy to determine suitable machine learning methods for a specific indoor positioning system. This helps to evaluate the performance of machine learning algorithms considering multiple criteria. During the experiments, UJllndoorLoc, KIOS and RFKON datasets are used from the positioning literature. The algorithms such as k-nearest neighbor, support vector machine, decision tree, naive bayes and bayesian networks are compared using these datasets. In addition to these, ensemble learning algorithms, namely adaboost and bagging, are utilized to improve the performance of these classifiers. As a conclusion, the test results for any specific dataset are reevaluated using the performance metrics such as accuracy, f-score and computation time, and a multi-criteria decision strategy is proposed to find the most convenient algorithm. The analytical hierarchy process is used for multi-criteria decision. To the best of our knowledge, this is the first work to select the proper machine learning algorithm for an indoor positioning system using multi-criteria decision strategy., Scientific and Technological Research Council of Turkey (TUBITAK) [1130024], This work is supported by The Scientific and Technological Research Council of Turkey (TUBITAK) under grant number 1130024.

The estimation of user position in indoor environment using WLAN technology based on Received Signal Strength (RSS) is becoming increasingly important in recent years. Various indoor positioning techniques are proposed in the literature. Fingerprint positioning technique is the most promising one that consists of radio frequency (RF) map construction and location estimation phases. Machine learning algorithms are used in the location estimation phase. Decision Tree algorithm is one of the most commonly applied ML algorithm that is used to infer user position by researchers. In this case study, we analyze decision tree algorithm parameters to find an optimal decision tree for indoor positioning system. The accuracy of this optimal tree is analyzed in the experiments.

Smart phones are widely used in indoor positioning and navigation systems. Thus, it is important to know the relative position of the smart phone on the user while calculating position with RSS based systems. The accuracy of position will change within this scope. In this study, an intermediary solution is proposed to find the position of the smart phone on the user. Several algorithms are evaluated for the feature selection process. The outputs of the process are estimated via three different classifiers, namely Naives Bayes, k-Nearest Neighborhood and C4.5 decision tree. k-Nearest Neighborhood algorithm achieves 96% accuracy with the output of Correlation Feature Selection.

Nowadays, robots have taken place of human beings in every area so the reliability of the robot has become important. Reliability analysis is applied to improve the reliability of imperfect robotic systems and to minimize the failures that may be occur in these systems. Overall reliability of the robot is obtained with applying the reliability analysis to all subsystems of robot. In this study, example of task completion study was made for robotic systems. As a starter task completion percentages are calculated with calculating failure rates under normal circumstances, after that with changing the ambient temperature and load carried task completion percentages are recalculating. As a result of study, effects of temperatures and loads to system reliability has been shown with tables and graphs.

International Symposium on Innovations in Intelligent SysTems and Applications (INISTA 2015) -- SEP 02-04, 2015 -- Madrid, SPAIN, WOS: 000380428200009, In fingerprinting-based indoor positioning systems, Received Signal Strength (RSS) values are collected at predetermined reference points to construct a fingerprint map. A well-established fingerprint database plays an important role in positioning, especially enhancing positioning accuracy. In literature, there are studies that consider only one type of measurements such as Wi-Fi or Bluetooth RSS, but these values are not sufficient alone to overcome the problems in dynamically changed environments. In order to deal with this, we propose a novel fingerprint database that contains both Wi-Fi and Bluetooth RSS values in addition to magnetic field measurements obtained from mobile devices. In addition to this, the proposed database also contains Wi-Fi, Bluetooth (BT) and Bluetooth Low Energy (BLE) RSS values obtained from preplaced sensor nodes in the experimental environment. The aims of this fingerprint database are to enhance accuracy, precision, and robustness of the location estimation system to dynamically changed environment and to satisfy researchers' needs who are deal with different problems in indoor positioning., Univ Autonoma Madrid, AIDA

23nd Signal Processing and Communications Applications Conference (SIU) -- MAY 16-19, 2015 -- Inonu Univ, Malatya, TURKEY, WOS: 000380500900178, The selection of appropriate classifier is of great importance in improving the positioning accuracy and processing time for indoor positioning. In this work, an extensive analysis is carried out to determine the most appropriate classification algorithm to solve the indoor positioning problem. KIOS Research Center dataset is used in the experimental work. Principal Component Analysis method is employed together with Ranker method to determine the best features. In the next stage, the performances of Naive Bayes, Bayesian Network, Multilayer Perceptron, K-Nearest Neighbor and J48 Decision Tree, which are widely preferred classification algorithms for indoor positioning studies, are analyzed on four distinct mobile phones. The results of the analysis reveal that J48 Decision Tree is superior to the other classification algorithms in terms of both processing time and accuracy., Dept Comp Engn & Elect & Elect Engn, Elect & Elect Engn, Bilkent Univ

International Symposium on Innovations in Intelligent SysTems and Applications (INISTA 2015) -- SEP 02-04, 2015 -- Madrid, SPAIN, WOS: 000380428200008, Fingerprinting based positioning is commonly used for indoor positioning. In this method, initially a radio map is created using Received Signal Strength (RSS) values that are measured from predefined reference points. During the positioning, the best match between the observed RSS values and existing RSS values in the radio map is established as the predicted position. In the positioning literature, machine learning algorithms have widespread usage in estimating positions. One of the main problems in indoor positioning systems is to find out appropriate machine learning algorithm. In this paper, selected machine learning algorithms are compared in terms of positioning accuracy and computation time. In the experiments, UJIIndoorLoc indoor positioning database is used. Experimental results reveal that k-Nearest Neighbor (k-NN) algorithm is the most suitable one during the positioning. Additionally, ensemble algorithms such as AdaBoost and Bagging are applied to improve the decision tree classifier performance nearly same as k-NN that is resulted as the best classifier for indoor positioning., Univ Autonoma Madrid, AIDA

Increasing population of elderly and people with disabilities constitute a huge demand for wheelchairs. Wheelchairs have an important place to improve the lives of and mobilize people with disabilities. Also, autonomous wheelchairs constitute a suitable research platform for academic and industrial researchers. In this study, a finite state machine-based high-level controller and a Kinect-based navigation algorithm have been developed for the Intelligent Wheelchair (ATEKS) that has high-tech control mechanisms, low-cost sensor equipment, and open source software (ROS, GAZEBO). Designed finite state machine and algorithm are tested in Gazebo.

Autonomous mobile vehicles are used in many applications to realize special tasks. These tasks involve obstacle avoidance, target reaching and/or tracking. Such vehicles include the use of artificial intelligence to assist the vehicle's operator. Fuzzy logic can be used in the design of an autonomous vehicle to improve the classical control mechanisms. Classical robot control/decision mechanisms can give imperfect results due to sensor compensation errors or calculation costs. These drawbacks can be eliminated by using a combined fuzzy inference. In this study, we have modified the mobile robot ATEKS, which is an intelligent wheelchair, by introducing three fuzzy inference systems to realize goal reaching, obstacle avoidance and a controller for combined behavior selection. Designed fuzzy control system has been implemented on Robot Operating System (ROS) under Ubuntu 12.04 operating system and tested under Gazebo simulation platform. Simulation results verified faithful behavior outputs of ATEKS.

Localization plays an important role in many applications. One of the positioning systems developed for indoor environments is ICKON that can calculate the position of an object at cm accuracy by using pure ultrasonic signals. In this study, the ICKON system is switched to wireless sensor network to reduce setup time, and a mobile application is developed to determine transmitter positions. Switching to wireless infrastructure also provides to use the existing wireless communication infrastructure in indoor environments.

Nowadays, location based applications increasingly used in many areas. Although outdoor navigation software applications are mature and widely used, improvements are required for indoor navigation. The applications such as guidance of disabled people, applications for security, visitor tracking, address mapping, the organization of services, automatic tourist guidance etc. are also needed for indoor environments. In this study, location-based navigation software has been developed for indoor environment. New plug-ins are added to navigation software, so that it can be used in Intelligent Wheelchair (ATEKS) systems, too. Thus, the software is specialized for autonomous robot application. The software is developed on Android platform.

Localization plays an important role in many applications such as robotic, signal processing and sensor networks. In the previous study, an indoor positioning system (ICKON) was developed. The ICKON uses only ultrasonic signals to calculate the position of mobile unit at cm level accuracy. For this accuracy, transmitter positions must be known in advance. These directly affect the mobile unit position calculation. In ICKON system transmitters' coordinates are determined by manual measurements. It needs long time depending on number of transmitters. In this study, a mobile application is developed for automatic transmitter position calculation. The mobile application uses Time of Arrival measurement and Trilateration method for the position calculation.

Localization is an important problem for robotics and mobile platforms. Although there is some globally accepted positioning systems (e.g., GPS, GLONASS) for outdoor environments, there is not such a system for indoor. In this study, an ultrasonic based positioning system (SESKON) is developed especially for indoor robotic applications. The SESKON uses only ultrasonic signals, so it differs from other positioning systems that are developed for indoor. The SESKON uses Time Difference of Arrival (TDOA) technique for position calculation. TDOA is preferred since it does not require clock synchronization between receiver and transmitters. Tests are conducted to show the effectiveness of the developed SESKON system.