CSEDU 2017 papers shortlisted for special issue in CCIS Springer series

In April 2017, SoftICE members Ottar L. Osen and Robin T. Bye presented two papers on educational research at CSEDU 2017. According to the CSEDU website, “[t]he very best papers presented at this event are selected by the conference and program chairs of the event based on a number of criteria that include the classifications and comments provided by the program committee members, the session chairs’ assessment and also the chairs’ global view of all papers included in the technical program.

Subsequently, the authors were informed that their papers “had been selected to be included in the Communications in Computer and Information Science (CCIS) series published by Springer. This book will include the updated and extended versions of a short list of selected papers from CSEDU 2017.

In order to have the extended version of the paper published in the Springer book, the authors have had to ensure that the extended versions had at least 30% of improvements. This work has now been completed, with both papers now including recent engineering student evaluation surveys at both the bachelor and master level at NTNU in Ålesund, updated literature reviews, and improved analyses.

The papers have just been submitted to Springer and will likely appear with the following titles (subject to change):

  • Ottar L. Osen and Robin T. Bye. Observations and Reflections on Teaching Electrical and Computer Engineering Courses. In Computer Supported Education: CSEDU 2017 – Revised Selected Best Papers, volume ZZZ of Communications in Computer and Information Science (CCIS), pages XX–YY. Springer-Verlag: Berlin Heidelberg, 2018.
  • Robin T. Bye. A Flipped Classroom Approach for Teaching a Master’s Course on Artificial Intelligence. In Computer Supported Education: CSEDU 2017 – Revised Selected Best Papers, volume ZZZ of Communications in Computer and Information Science (CCIS), pages XX–YY. Springer-Verlag: Berlin Heidelberg, 2018.

The paper abstracts are provided below.

Observations and Reflections on Teaching Electrical and Computer Engineering Courses

In this paper, the authors make a number of observations and reflections based their experiences over many years of teaching courses in electrical and computer engineering bachelor programmes. We present important aspects of attendance, lectures, group work, and compulsory coursework, and how these can be addressed to improve student learning. Moreover, we discuss how to facilitate active learning activities, focussing on simple in-classroom activities and larger problem-based activities such as assignments, projects, and laboratory work, and highlight solving real-world problems by means of practical application of relevant theory as key to achieving intended learning outcomes. Our observations and reflections are then put into a theoretical context, including students’ approaches of learning, constructive alignment, active learning, and problem-based versus problem-solving learning. Next, we present and discuss the results from two recent student evaluation surveys, one for senior (final-year) students and one for junior (first- and second-year students), and draw some conclusions. Finally, we add some remarks regarding our findings and point to future work.

Keywords: Active Learning; Problem-Solving Learning; Assessment; Engineering Pedagogy and Didactics.

A Flipped Classroom Approach for Teaching a Master’s Course on Artificial Intelligence

In this paper, I present a flipped classroom approach for teaching a master’s course on artificial intelligence. Traditional lectures from the classroom are outsourced to an open online course that contains high quality video lectures, step-by-step tutorials and demonstrations of intelligent algorithms, and self-tests, quizzes, and multiple-choice questions. Moreover, selected problems, or coding challenges, are cherry-picked from a suitable game-like coding development platform that rids both students and the teacher of having to implement much of the fundamental boilerplate code required to generate a suitable simulation environment in which students can implement and test their algorithms. Using the resources of the online course and the coding platform thus free up much valuable time for active learning in the classroom. These learning activities are carefully chosen to align with the intended learning outcomes, curriculum, and assessment to allow for learning to be constructed by the students themselves under guidance by the teacher. Thus, I perceive the teacher’s role as a facilitator for learning, much similar to that of a personal trainer or a coach. Emphasising problem-solving as key to achieving intended learning outcomes, the aim is to select problems that strike a balance between detailed step-by-step tutorials and highly open-ended problems.
This paper consists of an overview of relevant literature, the course content and teaching methods, recent evaluation reports and a student evaluation survey, results from the final oral exams, and a discussion regarding some limiting frame factors, challenges with my approach, and future directions.

Keywords: Flipped Classroom, E-Learning, Active Learning, Constructive Alignment, Problem-Solving, CodinGame, edX, C-4 Dynamite for Learning.}

Advertisements

Brice Assimizele defends PhD thesis

On 12 June 2017, Brice Assimizele defended his PhD thesis at Molde University College (MUC). The thesis is entitled Models and algorithms for optimal dynamic allocation of patrol tugs to oil tankers along the northern Norwegian coast and qualifies Dr. Assimizele to the Degree of Doctor of Philosophy in Logistics (Operations Research) –Specialized in Maritime Operations.

Dr. Assimizele was supervised by SoftICE member associate professor Robin T. Bye (NTNU Ålesund), associate professor Johan Oppen (MUC), and associate chair of research and professor of operations research Johannes O. Royset (Naval Postgraduate School, Monterey, CA, USA).

The very competent evaluation committee was headed by professor Lars Magnus Hvattum (MUC), with members professor and head of NTNU Oceans Ingrid Schjølberg (NTNU Trondheim), and associate professor in the Stewart School of Industrial & Systems Engineering Anton Kleywegt (Georgia Tech, Atlanta, USA).

About the thesis

Marine transportation of crude oil and petroleum products and its associated risk to the environment have increased significantly during the last decades. To safeguard the marine environment from potential oil spills and other damage from grounding of vessels, the Norwegian Coastal Administration (NCA) operates a center for vessel traffic service (VTS) in the town of Vardø in northeastern Norway.

The PhD research is conducted in close collaboration with the NCA, where the primary objective is to develop models and algorithms that optimally reduce the environmental costs from oil tankers grounding accidents. This includes optimal positioning of patrol tugs in a highly dynamic and stochastic environment. We propose a flexible and efficient decision support tool to the operators at the VTS, validated with historical events, that significantly reduce environmental risk associated with drifting vessels.

The methodological approach in this research could be applied to other search and rescue or emergency response related problems.

A fascinating life journey

Brice was born in Cameroon and raised by his grandmother, who, like many of Brice’s childhood friends, could not read or write. She understood the worth of education, however, and constantly helped pushing Brice into first completing school, before a bachelor’s degree in computer science at Université de Yaoundé 1 (Cameroon), a master’s degree in logistics management at UCSI University (Malaysia), and a master’s degree in industrial logistics: operations research at MUC. Now, with a PhD in logistics (operations research) at MUC, Brice has added another scalp to his belt in what must be considered a truly fascinating life journey.

During his PhD, Brice was employed by NTNU in Ålesund, where he had his own office and performed teaching duties along with his PhD work. His work is a continuation of the SoftICE project DRAMA.

The SoftICE lab congratulates Brice on this great achievement, and wish him the best of luck in his future endeavours!


Photos

The following photos are all courtesy Arild J. Waagbø, Panorama HiM. For more photos, please visit the Facebook page of Panorama.

Brice Assimizele getting ready for defense together with supervisors Johan Oppen and Robin T. Bye

Brice Assimizele.

Evaluation committee and opponents, Lars Magnus Hvattum, Ingrid Schjølberg, and Anton Kleywegt.

PhD defense by Brice Assimizele.

 

SoftICE presenting intelligent virtual prototyping at ECMS 2017

SoftICE members Robin T. Bye and Ibrahim A. Hameed will be presenting some recent research results on intelligent virtual prototyping of maritime winches in two scientific papers to be presented at the 31st European Conference on Modelling and Simulation (ECMS) 2017 in Budapest, Hungary, on 23–26 May. The papers are co-authored by the two abovementioned researchers together with SoftICE colleagues Ottar L. Osen and  Webjørn Rekdalsbakken, as well as Birger Skogeng Pedersen (Mechatronics Lab, NTNU):

  • Robin T. Bye, Ibrahim A. Hameed, Birger Skogeng Pedersen, and Ottar L. Osen. An intelligent winch prototyping tool. In Proceedings of the 31st European Conference on Modelling and Simulation (ECMS ’17), May 2017. Download pdf.
  • Ibrahim A. Hameed, Robin T. Bye, Birger Skogeng Pedersen, and Ottar L. Osen. Evolutionary winch design using an online winch prototyping tool. In Proceedings of the 31st European Conference on Modelling and Simulation (ECMS ’17), May 2017. Download pdf.

A Prezi presentation of the first paper is available here:

Interactive Prezi presentation: An Intelligent Winch Prototyping Tool (ECMS’17)The full papers are available for download here:  http://www.robinbye.com | Publications

The paper abstracts are provided at the end of this blog post.

Intelligent computer-automated design of cranes and winches

W build on our earlier work on intelligent computer-automated product design, where we have used methods from artificial intelligence (AI) such as genetic algorithms (GAs), particle swarm optimisation (PSO), and simulated annealing (SA) to optimise offshore crane design. Within a matter of only minutes, the algorithms were able to outperform the design of a real and delivered offshore crane with respect to some desired key performance indicators (KPIs). A human being would likely spend days or weeks to obtain the same results.

Generic and modular product optimization system.

Here, we focus on an intelligent winch prototyping tool (WPT):

Intelligent Winch Prototyping Tool (WPT)

We perform several test with various algorithms and are able to optimize a set of winch design parameter values that yield winch designs with suitable torque profiles:

Torque profiles for winch. The black profile has been optimized by means of a GA.

Abstract: An intelligent winch prototyping tool

In this paper we present a recently developed intelligent winch prototyping tool for ptimising the design of maritime winches, continuing our recent line of work using
artificial intelligence for intelligent computer-automated design of offshore cranes. The tool consists of three main components: (i) a winch calculator for determining key
performance indicators for a given winch design; (ii) a genetic algorithm that interrogates the winch calculator to optimise a chosen set of design parameters; and (iii) a web graphical user interface connected with (i) and (ii) such that winch designers can use it to manually design new winches or optimise the design by the click of a button. We demonstrate the feasibility of our work by a case study in which we improve the torque profiles of a default winch design by means of optimisation. Extending our generic and modular software framework for intelligent product optimisation, the winch calculator can easily be interfaced to external product optimisation clients by means of the HTTP and WebSocket protocols and a standardised JSON data format. In an accompanying paper submitted concurrently to this conference, we present one such client developed in Matlab that incorporates a variety of intelligent algorithms for the optimisation of maritime winch design.

Abstract: Evolutionary winch design using an online winch prototyping tool

This paper extends the work of a concurrent paper on an intelligent winch prototyping tool (WPT) that is part of a generic and modular software framework for intelligent computer-automated product design. Within this framework, we have implemented a Matlab winch optimisation client (MWOC) that connects to the WPT and employs four evolutionary optimisation algorithms to optimise winch design. The four algorithms we employ are (i) a genetic algorithm (GA), (ii) particle swarm optimisation (PSO), (iii) simulated annealing (SA), and (iv) a multi-objective optimisation genetic algorithm (MOOGA). Here, we explore the capabilities of MWOC in a case study where we show that given a set of design guidelines and a suitable objective function based on these guidelines, we are able to optimise a particular winch design with respect to some desired design criteria. Our research has taken place in close cooperation with two maritime industrial partners, Seaonics AS and ICD Software AS, through two innovation and research projects on applying artificial intelligence for intelligent computer-automated design of maritime equipment such as offshore cranes and maritime
winches.

More information

We have previously presented some details of our work on intelligent virtual prototyping of cranes and winches in earlier blog posts:

Acknowledgements

The SoftICE lab at NTNU in Ålesund wishes to thank ICD Software AS for their contribution in the software development process, and Seaonics AS for providing
documentation and insight into the design and manufacturing process of offshore cranes. We are also grateful for the support provided by Regionalt Forskningsfond
(RFF) Midt-Norge and the Research Council of Norway through the VRI research projects Artificial Intelligence for Crane Design (Kunstig intelligens for krandesign
(KIK)), grant no. 241238, and Artificial Intelligence for Winch Design (Kunstig intelligens for vinsjdesign (KIV)), grant no. 249171.

Collaboration?

Parties interested in research collaboration, testing our software, or more information are encouraged to contact us.

The SoftICE Lab

NTNU Ålesund students win prestigious automation engineering award on USVs for aqua farm inspection

Better late than never!

In February 2017, bachelor students of automation engineering at NTNU Ålesund, Albert Havnegjerde, Vegard Kamsvåg og Sveinung Liavaag, won the prestigious Norwegian national award for the best bachelor thesis 2016 on automatic control given by the Norwegian Society for Automatic Control (NFA).

Jury member Rune Volden from Ulstein Power & Control hands over the award to NTNU Ålesund students Sveinung Liavaag and Albert Havnegjerde. Vegard Kamsvåg was unable to attend. Image courtesy: NFA

The students also co-wrote a paper based on their work together with SoftICE members Robin T. Bye and Ottar L. Osen (student supervisor) that was presented at IEEE Techno-Ocean 2016 and subsequently published in the proceedings:

  • Ottar L. Osen, Albert Havnegjerde, Vegard Kamsvåg, Sveinung Liavaag, and Robin T. Bye. A Low Cost USV for Aqua Farm Inspection. In Proceedings of IEEE Techno-Ocean ’16, pages 291–298, October 2016.

In their work, the students employed rapid prototyping to develop a low cost (~2000 EUR) remotely controlled unmanned surface vessel (USV) intended for inspection of aqua farms whilst incorporating a dynamic positioning (DP) system.

This work was partly financed by an internal educational project called Research-based and Innovation-driven Learning through Final Year Projects (Forskningsbasert og innovasjonsdrevet læring gjennom avsluttende oppgave – FILA).

The full paper and the conference presentation is available for download here: www.robinbye.com | Publications

Abstract: A Low Cost USV for Aqua Farm Inspection

This paper describes the rapid prototyping of a low cost remotely controlled unmanned surface vessel (USV) intended for inspection of aqua farms. There is an increased focus on inspection of ocean-based aqua farms due to three major challenges: escaping fish, sea lice, and algae. Escaping fish may bring diseases to other fish or interbreed with wild fish and damage their gene material. Sea lice is a parasite that may seriously damage the fish, lower its food quality, and if not treated, can spawn and multiply into an epidemic. Finally, algae blooms may lower oxygen levels and kill the fish. To proactively counter these challenges, aqua farm operators need to regularly inspect the fish cages for holes, the water for algae, and the fish for sea lice. Modern ocean-based aqua farms are usually constructed with two rows of sea cages separated by a gangway in the middle, often with a small operation and machinery building at one end. Staff visually inspect the cages from above and from the nearside by walking up and down the gangway. Inspection of the outer side of a cage will normally require a boat with a human inspector on board, whereas subsea inspection will normally require a human diver. Here, we propose a USV design solution for this kind of inspection that provides the aqua farm operator with a remotely controlled unmanned boat and subsea video feed. A working prototype has been designed in less than six months and successfully tested at sea.

Index Terms—USV; ROV; dynamic positioning; low cost; commercial off-the-shelf; rapid prototyping; aquaculture.

Undervisningsseminar 27. mars

Undervisningsseminar

Automagisk sensur, adaptivt læringsverktøy, og aktiv læring

mandag 27. mars 2017 kl. 10-12

Auditorium Nørvasundet, Hovedbygningen, NTNU i Ålesund

Software and Intelligent Control Engineering (SoftICE) Laboratory ved Institutt for IKT og realfag (IIR) inviterer med dette til åpent seminar om moderne undervisning i høyere utdanning.

Program

10.00 Åpning v/Annik Magerholm Fet, viserektor, NTNU

10.05 System for automatisk individuell faglig begrunnelse og tilbakemelding v/Omid Mirmotahari, førsteamanuensis, Studielaben, Institutt for informatikk, UiO

11.00 Adaptivt læringsverktøy for matematikk v/Siebe van Albada, studieprogramleder simulering og visualisering, IIR, NTNU

11.20 Modelleringsverktøy for dingser v/Adrian Rutle, studiekoordinator informasjonsteknologi, Institutt for data- og realfag, HVL

11.40 Aktiv læring i matematikk v/Hans Georg Schaathun, professor, SoftICE Lab, IIR, NTNU

12.00 Avslutning

Hovedinnlegget holdes av Omid Mirmotahari, som har vunnet en rekke priser for undervisning, forskning og formidling. Nylig har han også figurert i media med innovative “automagiske” system for automatisk individuell faglig begrunnelse og tilbakemelding:

For mer informasjon, kontakt Robin T. Bye på robin.t.bye@ntnu.no eller +47 40082880.

Invitasjon til seminar: Undervisningsseminar2017.pdf

SoftICE presents educational research at CSEDU 2017

SoftICE members Ottar L. Osen and Robin T. Bye and will be presenting two educational research papers at the 9th International Conference on Computer Supported Education (CSEDU 2017) in Porto, Portugal on 21–23 April:

  • Ottar L. Osen and Robin T. Bye. Reflections on teaching electrical and computer engineering courses at the bachelor level. In Proceedings of the 9th International Conference on Computer Supported Education — Volume 2: CSEDU (CSEDU ’17), pages 57–68. INSTICC, SCITEPRESS, April 2017. Download pdf. View Prezi.
  • Robin T. Bye. The teacher as a facilitator for learning: Flipped classroom in a master’s course on artificial intelligence. In Proceedings of the 9th International Conference on Computer Supported Education — Volume 1: CSEDU (CSEDU ’17), pages 184–195. INSTICC, SCITEPRESS, April 2017. Download pdf. View Prezi.

The full papers and other work is available for download here: http://www.robinbye.com | Publications

The paper abstracts are provided below.

Reflections on teaching electrical and computer engineering courses at the bachelor level

This paper reflects on a number of observations the authors have made over many years of teaching courses in electrical and computer engineering bachelor programmes.
We suggest various methods and tips for improving lectures, attendance, group work, and compulsory coursework, and discuss aspects of facilitating active learning, focussing on simple in-classroom activities and larger problem-based activities such as assignments, projects, and laboratory work. Moreover, we identify solving real-world problems by means of practical application of relevant theory as key to achieving intended learning outcomes. Our observations and reflections are then put into a theoretical context, including students’ approaches of learning, constructive alignment, active learning, and problem-based versus problem-solving learning. Finally, we present and discuss some recent results from a student evaluation survey and draw some conclusions.

The teacher as a facilitator for learning: Flipped classroom in a master’s course on artificial intelligence

In this paper, I present a flipped classroom approach for teaching a master’s course on artificial intelligence. Traditional lectures in the classroom are outsourced to an open online course to free up valuable time for active, in-class learning activities. In addition, students design and implement intelligent algorithms for solving a variety of relevant problems cherrypicked from online game-like code development platforms. Learning activities are carefully chosen to align with intended learning outcomes, course curriculum, and assessment to allow for learning to be constructed by the students themselves under guidance by the teacher, much in accord with the theory of constructive alignment. Thus, the teacher acts as a facilitator for learning, much similar to that of a personal trainer or a coach. I present an overview of relevant literature, the course content and teaching methods, and a recent course evaluation, before I discuss some limiting frame factors and challenges with the approach and point to future work.

Opening of Telenor-NTNU AI-Lab and its first Hackathon

The Telenor-NTNU AI-Lab was officially opened on 8 March 2017, when several prominent guests, including Norwegian Minister of Trade and Industry Monica Mæland, Norwegian Minister of Culture Linda Hofstad Helleland, SINTEF CEO Alexandra Bech Gjørv, and Telenor CEO Sigve Brekke, amongst others, joined NTNU rector Gunnar Bovim and head of the Department of Computer Science at NTNU, Letizia Jaccheri for celebration.

Celebrity guests meets Inge, one of the SoftICE Lab’s social robots.

About the AI-Lab

The Telenor-NTNU AI-Lab is a joint lab for research in Artificial Intelligence, Machine Learning, and Big Data Analytics. The lab was established in 2016, and has been formally operative from January 1st, 2017. It is hosted by the Department of Computer Science. The lab will conduct fundamental ML research, including theory and method development, as well as application-oriented research at a high international level. Lab facilities will also be available for other research groups within the Faculty of Information Technology and Electrical Engineering (IE) doing ML research, for NTNU more generally, and for external cooperating partners.

Telenor-NTNU AI-Lab was established as part of Telenor’s vision to help Norway deal with the challenges of an increasing digitized society. The SoftICE Lab intends to be contribute to reaching this goal.

Hackathon

The AI-Lab will host its very first hackathon on the weekend 17-18 March both in Trondheim and at the SoftICE Lab on Campus Ålesund.

If you are a student or employee at NTNU i Ålesund, please contact SoftICE member Ibrahim A. Hameed on Facebook or by email more information.

The hackathon will take place both in Trondheim and on Campus Ålesund. There will be served pizza on Friday and breakfast and lunch on Saturday.

Photos from the opening of the AI-Lab

NTNU_AI-lab-4495