Wyniki wyszukiwania - BazTech

1

The Machine Learning Method of PIDVCA

Li L., Wang X., Chen G.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2020

|

Vol. 14. no. 3

533--540

EN

Building a dynamic collision knowledge base of self-learning is one of the core contents of implementing "personified intelligence" in Personifying Intelligent Decision-making for Vessel Collision Avoidance (short for PIDVCA). In the paper, the machine learning method of PIDVCA combined with offline artificial learning and online machine learning is proposed. The static collision avoidance knowledge is acquired through offline artificial learning, and the isomeric knowledge representation integration method with process knowledge as the carrier is established, and the Dynamic collision avoidance knowledge is acquired through online machine learning guided by inference engine. A large number of simulation results show that the dynamic collision avoidance knowledge base constructed by machine learning can achieve the effect of anthropomorphic intelligent collision avoidance. It is verified by examples that the machine learning method of PIDVCA can realize target perception, target cognition and finally obtain an effective collision avoidance decision-making.

2

A suggestion of an application of blended learning in MET through a harmonized STCW Model

Pipchenko O. D., Kovtunenko D.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2020

|

Vol. 14. no. 3

545--548

EN

The article addresses the problem of modern maritime education and training in the perspective of computer technologies development, especially the internet. Computer-based training is being a standard for the maritime industry for almost 25 years, but there is still no unified approach on the use of this teaching method in MET. Authors suggest to open a conversation on harmonization of standards regarding CBT in the industry and wider implementation of this teaching method into STCW training process via a blended learning approach, where theoretical part of the course may be taken online, while the practical part is taken in the training facility.

3

A Study of Correlation between Fishing Activity and AIS Data by Deep Learning

Shen K. Y., Chu Y. J., Chang S. J., Chang S. M.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2020

|

Vol. 14. no. 3

527--531

EN

Previous researches on the prediction of fishing activities mainly rely on the speed over ground (SOG) as the referential attribute to determine whether the vessel is navigating or in fishing operation. Since more and more fishing vessels install Automatic Identification System (AIS) either voluntarily or under regulatory requirement, data collected from AIS in real time provide more attributes than SOG which may be utilized to improve the prediction. To be specific, the ships' trajectory patterns and the changes in course become available and should be considered. This paper aims to improve the accuracy in the identification of fishing activities. First, we do feature extraction from the AIS data of coastal waters around Taiwan and build a Recurrent Neural Network (RNN) model. Then, the activity data of fishing vessels are divided into fishing and non-fishing. Finally, based on the testing by feeding various fishing activity data, we can identify the fishing status automatically.

4

Developing generative adversarial nets to extend training sets and optimize diiscrete actions

Zhang R. L., Furusho M.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2019

|

Vol. 13. no. 4

875--880

EN

This study proposes the use of generative adversarial networks (GANs) to solve two crucial problems in the unmanned ship navigation: insufficient training data for neural networks and convergence of optimal actions under discrete conditions. To achieve smart collision avoidance of unmanned ships in various sea environments, first, this study proposes a collision avoidance decision model based on a deep reinforcement learning method. Then, it utilizes GANs to generate enough realistic image training sets to train the decision model. According to generative network learning, the conditional probability distribution of ship maneuvers is learnt (action units). Subsequently, the decision system can select a reasonable action to avoid the obstacles due to the discrete responses of the generated model to different actions and achieve the effect of intelligent collision avoidance. The experimental results showed that the generated target ship image set can be used as the training set of decision neural networks. Further, a theoretical reference to optimize the optimal convergence of discrete actions is provided.

5

Wiedza studentów Informatyki Politechniki Lubelskiej o platformach e-lerningowych

Borys M., Łukasik E.

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2014

|

nr 2

13--15

PL

W latach 2011-2012 realizowany był na Wydziale Elektrotechniki i Informatyki (WEiI) Politechniki Lubelskiej projekt POKL „Absolwent na miarę czasu” skierowany do studentów studiów magisterskich kierunku Informatyka. Wśród celów realizowanego projektu ważne miejsce zajmuje utworzona platforma e-learningowa. W niniejszym artykule omówiono obecny stan wykorzystania platformy, zaprezentowano wyniki badań przeprowadzonych wśród studentów dotyczących: poziomu wiedzy studentów o nauczaniu na odległość, oceny stworzonej platformy oraz przydatności materiałów na niej umieszczonych.

EN

In 2011-2012 at the Faculty of Electrical Engineering and Computer Science the project called "Graduate of our time" aimed for students of Master studies in Computer Science is realized. One of the most important objectives of the project is implementation of e-learning platform. The article discusses the current state of the use of the platform, presents the results of the survey conducted among students regarding: the level of students’ knowledge of distance learning, assessment of created platform as well as assessment developed learning resources.

6

Metoda early stopping w optymalizacji neuronowych estymatorów zmiennych stanu Układu napędowego z połaczeniem elastycznym

Kamiński M., Orłowska-Kowalska T.

Prace Naukowe Instytutu Maszyn, Napędów i Pomiarów Elektrycznych Politechniki Wrocławskiej. Studia i Materiały

|

2009

|

Vol. 63, nr 29

202-216

PL

W artykule opisano neuronowe estymatory zmiennych stanu napedu elektrycznego z połaczeniem spreżystym. Sieci neuronowe zastosowane zostały do odtwarzania momentu skretnego napedu oraz prędkości silnika obciażajacego. Przeprowadzona została analiza wpływu metody wczesnego zatrzymania procesu uczenia (early stopping) na jakosc odtwarzania zmiennych stanu. Charakterystyczna cecha tej metody jest brak ingerencji w strukture sieci neuronowej oraz algorytmu treningowego. Zastosowanie tej procedury na etapie projektowania estymatorów neuronowych ma na celu zwiekszenie dokładności odtwarzania poszczególnych sygnałów oraz uzyskanie odpornosci na zakłócenia zewnetrzne (zmiany parametrów obiektu, załaczanie momentu obciażenia). Przedstawiono wyniki badań symulacyjnych i eksperymentalnych ilustrujacych efektywnosc proponowanej metody optymalizacji sieci neuronowych.

EN

In the paper neural estimators of state variables of the electrical drive with elastic coupling are presented. Neural network were applied for the estimation of the shaft torque and angular speed of the load machine. An analysis of an early stopping method under training process of neural networks on the estimation quality was performed. The characteristic feature of this method is a lack of influence on the neural network structure and training algorithm. Application of this procedure during the design stage of neural networks improves significantly the estimation quality and robustness (to changes of the system parameters, changes of the load torque) of the proposed neural estimators of the two-mass system. Simulation and experimental tests illustrating the effectiveness of the proposed method are demonstrated.

7

Analiza wpływu sposobu przygotowania danych uczących dla sieci neuronowych na jakość odtwarzania zmiennych stanu napędu dwumasowego

Kamiński M.

Prace Naukowe Instytutu Maszyn, Napędów i Pomiarów Elektrycznych Politechniki Wrocławskiej. Studia i Materiały

|

2009

|

Vol. 63, nr 29

426-438

PL

W artykule opisano neuronowe estymatory zmiennych stanu napędu elektrycznego z połączeniem sprężystym. Sieci neuronowe zastosowane zostały do odtwarzania momentu skrętnego napędu oraz prędkości silnika obciążającego. Przeprowadzona została analiza wpływu wprowadzania szumu do sygnałów wykorzystywanych w procesie treningu sieci neuronowych (jittering) na jakość odtwarzania zmiennych stanu. Charakterystyczną cechą tej metody jest brak ingerencji w strukturę sieci neuronowej oraz algorytmu treningowego. Zastosowanie tej procedur na etapie projektowania estymatorów neuronowych ma na celu zwiększenie dokładności odtwarzania poszczególnych sygnałów oraz uzyskanie odporności na zakłócenia zewnętrzne Przedstawiono wybrane wyniki badań symulacyjnych i eksperymentalnych ilustrujące skuteczność tej metody uczenia estymatorów neuronowych dla napędu dwumasowego.

EN

In the paper neural estimators of state variables of the electrical drive with elastic coupling are presented. Neural network were applied for the estimation of the shaft torque and angular speed of the load machine. An analysis of a jittering method under training process of neural networks on the estimation quality was performed. The characteristic feature of this method is a lack of influence into the neural network structure or training algorithm. Application of this procedure during the design stage of neural networks improves significantly the estimation quality and robustness (to changes of the system parameters, changes of the load torque) of the proposed neural estimators of the two-mass system. Simulation and experimental tests illustrating the effectiveness of the proposed method are demonstrated.

8

Metody uczenia sieci neuronowej Hopfielda

Matusik R.

Pomiary Automatyka Kontrola

|

2009

|

R. 55, nr 7

521-523

PL

W artykule przedstawione zostały od strony teoretycznej i porównane od strony praktycznej różne metody uczenia sieci neuronowej Hopfielda. Oprócz znanej i powszechnie stosowanej reguły Hebba, przedstawione zostały modyfikacje tej metody. W celu porównania reguł uczenia sieci Hopfielda napisana została specjalna aplikacja, w której zaimplementowane zostały przedstawione w artykule metody. Regułą najlepiej rozpoznającą zapamiętane wzorce okazała się metoda pseudoinwersji

EN

The Hopfield neural network can have many applications, such as approximation, compression, association, steering or patterns recognition. If the neural network is used for association, it is an associative memory. This task consists in original patterns recognition even when the Hopfield neural network is cued with distorted patterns. In this paper various learning methods for the Hopfield neural network are presented from the theoretical point of view and they are compared from the practical point of view. Besides the well known and generally used Hebb rule, there are presented its modifications as well. In order to compare the learning methods for the Hopfield neural network, a special application in which there are implemented the methods described in the paper is written. Section 2 contains the Hopfield neural network model, the Hopfield neural network definition and the neural network general schematic. There is also de-scribed the activation function used for testing the Hopfield neural network. Section 3 gives various Hopfield network learning rules, such as the original Hebb method, its modifications, the Oja rule and pseudoinversion rule. In Section 4 the testing process and its results are presented. The main task of this neural network is patterns recognition. The Hopfield neural network stored 10 patterns. Each of the stored patterns had 35 neurons. Then the neural network was cued with distorted patterns. The tests proved that the pseudoinversion rule recognized the patterns in the best way.

9

Business English as an intellectual bridge -management of the syncretic case study method, organizational changes in education management and blended learning for non-native Business English and management sciences students in an intercultural environmen

Odrakiewicz P.

Organizacja i Zarządzanie : kwartalnik naukowy

|

2009

|

nr 3

59-77

EN

Business English can be viewed as a core cross-cultural competence in today's globalized village. It is also an intellectual bridge for better understanding. This paper is based on my own teaching and education management experience of more than fifteen years in the United States, Canada and Poland where I have taught Business English and management. Instructors of Business English as a second language (ESL) and management sciences could greatly enhance their students' learning by employing the case-study method and e-learning in tandem. I outline the characteristics of what I call the syncretic case study method which is a blend of two case study approaches, the Western Ontario University and Harvard methods. Business English as a central component of curriculum must take into act the interconnected and multicultural world.

PL

Business English może być postrzegany jako kluczowa kompetencja w globalnej wiosce. Jest on również intelektualnym pomostem lepszej komunikacji w biznesie. Poniższy artykuł jest podsumowaniem mojego ponad 15-letniego doświadczenia w nauczaniu i zarządzaniu edukacją w Kanadzie, USA i Polsce. Artykuł przedstawia zalety zastosowanie metody synkretycznej, będącej wypadkową metod nauczania za pomocą case studies przez uniwersytety Harvarda i Western Ontario. Business English jako centralna część programu komunikacji w organizacji musi brać pod uwagę powiązany i multikulturowy świat.

10

The role of interaction between students in the process of discovering the regularity

Pytlak M.

Scientific Issues of Jan Długosz University in Częstochowa. Mathematics

|

2007

|

Vol. 12

367-374

EN

In teaching mathematics, interactions between the teacher and the student and among students play a vital role. Through making students formulate and defend their points of view we develop in them their self-control. Thanks to it during solving problem a child is more responsible and conscious of what s/he does. Necessity of verbalization of executing activities and explanation of using procedures show that pupils are able to notice new things. The verbalization forces to look at the own work from a different perspective. In this paper I present a part of my research concerning discovering the regularity by 9-years old children. In this research I focused on mental process and interaction between the students.

11

Learning methods in ad-hoc networks: a review

Kudelski M., Pacut A.

Prace Naukowe Politechniki Warszawskiej. Elektronika

|

2007

|

z. 160

153-163

EN

The paper presents a comprehensive review of learning methods used to solve various problems in ad-hoc networks. The learning methods are classified according to learning mechanisms and problems solved. Nine representative approaches are discussed in more detail.

12

Comparative study of learning methods for artificial neural network

Tiliouine H.

Pomiary Automatyka Kontrola

|

2007

|

R. 53, nr 4

117-121

EN

The paper presents a comparative study of various learning methods for artificial neural network. The methods are: the backpropagation BP, the recursive least squares RLS, the Zangwill's method ZGW and the method based on evolutionary algorithm EA. The study consists of evaluating the learning effectiveness of these methods and selecting the most efficient one to be used in the designing of an adaptive neural voltage controller for a synchronous generator.

PL

W artykule przedstawiono wyniki badań porównawczych metod uczenia sieci neuronowych takich jak: metoda propagacji wstecznej błędów, rekurencyjna metoda najmniejszych kwadratów, metoda Zangwill'a, metoda algorytmów ewolucyjnych. Celem tych badań jest dobieranie najefektywniejszej metody uczenia do projektowania adaptacyjnego neuronowego regulatora napięcia generatora synchronicznego.