Kastelan, I., Lopez, B. J. R., Artetxe, G. E., Piwinski, J., Barak, M., & Temerinac, M. (2014). E2LP: A unified embedded engineering learning platform. Microprocessors and Microsystems, 38(8), 933-946.
E2LP: A Unified Embedded Engineering Learning Platform Ivan Kastelan1, Jorge R. Lopez Benito2, Enara Artetxe Gonzalez2, Jan Piwinski3, Moshe Barak4, Miodrag Temerinac1 1 University of Novi Sad, Faculty of Technical Sciences, Fruskogorska 11, 21000 Novi Sad, Serbia CreativiTIC INNOVA S.L Centro Tecnológico de La Rioja, Avenida de Zaragoza 21, 26006 Logroño, Spain 3 Industrial Research Institute for Automation and Measurements (PIAP), Al. Jerozolimskie 202, 02-486 Warsaw, Poland 4 Ben-Gurion University of the Negev, P.O.B. 653 Beer-Sheva 8410501, Israel
[email protected],
[email protected],
[email protected],
[email protected],
[email protected],
[email protected] 2
Abstract— The main idea behind this project is to provide a unified platform which will cover a complete process for embedded systems learning. A modular approach is considered for skills practice through supporting individualization in learning. This platform shall facilitate a novel development of universal approach in creative learning environment and knowledge management that encourage use of ICT. New learning model is challenging the education of engineers in embedded systems design through real-time experiments that stimulate curiosity with ultimate goal to support students to understand and construct their personal conceptual knowledge based on experiments. In addition to the technological approach, the use of cognitive theories on how people learn will help students to achieve a stronger and smarter adaptation of the subject. Applied methodology will be evaluated from the scientific point of view in parallel with the implementation in order to feedback results to the R&D. Keywords-technology enhanced learning, embedded systems, learning platform, engineering education
I.
INTRODUCTION
Embedded systems are invisible electronics and corresponding software that bring intelligence to objects, processes and devices. The main challenge in engineering education for embedded systems is a complex interdisciplinary approach which includes: understanding of various systems based on different technologies and system solution optimizations. The significance of laboratory work in electrical engineering education has been widely recognized. It provides engineering education personnel to transform passive listener’s students into active learners, thus stimulating students to actively participate in the learning process [1]. Moreover, knowledge obtained through laboratory hands-on experience has proven to be more profound and more lasting. Increased role of computer based embedded systems in various industrial applications has produced a growing need for embedded system engineers. As of today, the job market is very competitive for highly qualified electronics systems engineers [2]. Furthermore, the European Centre for the Development of Vocational Training (Cedefop) in its publication "Skills supply and demand in Europe, Medium-term forecast up to 2020" [3] predicts that increase of 10.7% in new engineering positions across all disciplines in Europe will be reached up to 2020, compared to the 2010 level. Similar projections are available for USA by the Bureau of Labor Statistics, U.S. Department of Labor in “Occupational Outlook Handbook, 2010-11 Edition” [4] and Asia given in "Employment Outlook Asia: Focus on China and India" written by Mary Anne Thompson [5]. Consequently, many technical faculties have put more emphasis on embedded systems learning by introducing a number of active learning laboratory-based courses [1] and [6]-[10]. A typical approach in the field of computer engineering education is to have a group of courses for a specific computer engineering topic. Usually, these courses comprise a set of laboratory assignments which are performed using different hardware platforms for each particular course, thus introducing high cost requirements for learning equipment. Further, the dynamics necessary to address industry needs have put teaching personnel under significant pressure to adequately design the laboratory environment for the courses, since updates of the learning environment in an embedded systems course are necessary on a regular basis to keep it relevant. Having this in mind, it is not hard to imagine technical faculties ending up with a set of different and often inconsistent laboratory setups and teaching platforms used throughout the curriculum. As a result, efficiency of laboratory work in embedded systems learning usually suffers from introduced overhead (30%) in both the time and the effort necessary to get students familiar with hardware platforms and software tools for each course.
The approach in this project targets the lab education efficiency with the idea to use a single comprehensive platform for the complete curriculum. The main intention is to make the educational process more efficient and to introduce more interaction between the education and further embedded system research and development, which facilitates an optimal solution for a specific problem. The main E2LP objective is to efficiently educate future engineers capable of coping with current challenges in real-time embedded computer engineering field. It will further provide a learning environment that moves focus from hardware to software and encourages learning of the embedded systems, but without compromising a knowledge related to the hardware design. Additionally, by introducing a unified platform on a European level, easier interaction in collaborative-education between different scientific institutions will be enabled. Furthermore, the E2LP platform shall introduce a flexible and extendable learning environment for new-coming technologies in embedded systems, thus providing a long lasting educational solution for academia. The rest of the paper is organized as follows: section 2 summarizes the requirements for the learning platform. Section 3 provides an overview of the concept of the unified platform. Section 4 discusses the central part of the platform – the Base Board. Finally, section 5 gives some concluding remarks. II.
RELATED WORK
Many electrical engineering faculties have put more emphasis on embedded systems learning by introducing a number of active learning laboratory-based courses. Bindal, Mann, Ahmed and Raimundo [7] developed a senior-level undergraduate system-onchip (SoC) course that emphasizes SoC design methods and hardware-software co-design techniques. The course uses a "real world" design project as the teaching vehicle. Foist, Grecu, Ivanov and Turner [8] presented a reference design and tutorial for an embedded PowerPC subsystem core with user logic in a field-programmable gate array (FPGA). Hamblen [9] gives an approach to teaching in an embedded systems design course which uses a low-cost SoC computer and a real-time operating system (RTOS). The industry needs for embedded engineers has also made an influence on applied teaching methodology. For example Velez and Sevillano [10] have proposed project based approach to teaching digital hardware design with clear analogy to the industry project flow. The students must follow the entire design process: from interpretation of written specification to the delivery of functionally verified code. The goal of all engineering universities in the world is to produce high-qualified graduates capable to challenge engineering tasks. In particular, for embedded engineering it is important that graduates embraced software and hardware components of embedded systems in a variety of applications. IEEE Computer Society and Association for Computer Engineering have proposed guidelines for designing an undergraduate curriculum in computer engineering [11]. University of Alabama, USA has integrated the concepts [11] into their embedded engineering curriculum, which is reported by Ricks et al. in [12]. This framework is designed for educational needs of local industry in the state of Alabama and is consisted of sequence of 7 courses:
Digital Logic,
Microcomputers,
Area Electives,
Digital Systems Design,
Embedded Systems,
Computer Architectures,
Capstone Design.
The curriculum organization given in [12] is a very illustrative example of a typical embedded engineering education process consisted of hardware and software components, with included team-oriented design projects. However, this particular example is slightly more oriented towards hardware designs and without digital signal processing topics, although they are recommended in [11]. Framework example for teaching digital signal processing (hardware oriented) is given by Hall and Anderson in [13]. Another example of embedded systems curriculum is given in [14]. Applied learning model in this case emphasizes significance in the laboratory assignments. The cognitive learning in the laboratory is used for gaining a better understanding of teaching subjects. Pierre et al. in [15] are presenting a nice introductory course in electrical engineering with usage of variety of topics including: circuits, electronics, digital logic, microprocessors, communications, digital signal processing and power. Although each of this topics is complex enough for dedicated course, in this example students are taken through all important subjects in embedded engineering through hands-on examples within one single course at their first year of studies. Such one “embedded system overview” approach provides students a nice perspective what is computer engineering and motivates their further studies in accordance with their interests and gained experience.
The success of these examples within the student population is encouraging the design of multidisciplinary active learning platform, as suggested by Ravel and McDermott in [16]. These results and feedback received from global faculty and students across North America, Europe and Asia identified barriers they faced in adopting a laboratory-based active learning pedagogy in their embedded systems courses [17].
Too much time needed to learn a new tool for every class. Tools are complex, only a subset is needed.
Long cycle time to acquire parts and fabricate prototypes.
Too much time to guide disparate design projects.
Design tools may need a lot of support. Many issues in software installation, revision, licensing and conflicts.
Facilities for design and prototyping are hard to acquire & maintain, very little reuse across courses.
Based on these observations, and built on comprehensive industrial experience in platform-based engineering, a modular common design platform was developed and introduced to curriculum in the University of Novi Sad, Faculty of Technical Sciences [18]. Besides the University of Novi Sad, the platform is also introduced in the number of universities worldwide, with their feedback presented in [17]. The overview of students’ perspective in usage of this platform for digital design course is given in [19] and initial set of exercises developed for this platform in [20]. The E2LP project intends to overcome existing learning overhead through innovative engagement in the learning process management. Augmented reality interface (AR) was initially developed on the basis of Virtual Reality technologies. It intends to augment the user’s perception by integrating virtual objects into a real environment in real time [21]-[23]. Various types of augmentation are currently proposed: visual, audio, haptic, etc. Research in the field of AR for learning and training is essentially technology-driven [24]. Currently there are only few real educational AR-based applications and there is not much user feedback on AR utility for learning processes. Nevertheless, there has been some speculation on the potential of AR to assist learning by facilitating learners’ informational access during technical operations on invisible machine parts, reducing the error likelihood by introducing “virtual reminders”, enhancing motivation, reducing the volume of paper-based technical documentation and introducing the possibility of on-the-job training [25]. These examples show that AR is essentially regarded as an informational training aid through visual and audio computer input. From a learning point of view, an AR application could only be attractive if it were built upon real needs, with a special emphasis on userand learner-centered design [26]. Bearing this aspect in mind and taking into consideration the fact that a learner-centered approach is often absent from the design of AR technologies, we will introduce these elements in the design of the AR system in this project. III.
EXPERIENCE-BASED EDUCATIONAL REQUIREMENTS FOR A LEARNING PLATFORM
The empirical evaluation of students’ and teachers’ perception of current courses in embedded computer engineering curriculum was done in the University of Zagreb and University of Novi Sad. Below the results from the University of Zagreb, Faculty of Electrical Engineering and Computing and University of Novi Sad, Faculty of Technical Sciences are presented. At the University of Zagreb, the numbers of students who responded to the questionnaire were 80 for the Digital Logic and Computer Architecture 1 course, 15 for the Computer Architecture 2 course, and eight for the Multimedia Architecture and Systems course (a total of 103 answers). At the University of Novi Sad, the number of students who responded to the questionnaire was 30 for the Digital System Design. Additionally, 13 teachers responded to the teachers’ questionnaire in five course categories at the University of Novi Sad. To summarize the students' answers, we calculated the percentage of students who marked 'Very much' or 'Much' for the six items for each of the five selected courses. The outcomes are presented in Fig. 1. It may be seen in Fig. 1 that 71% of the students marked 'Very much' or 'Much' for the item "We deal with basic exercises, drills and practice," which are basic assignments in the taxonomy (scale) we present later in this document.
Figure 1. Students’ (upper) and teachers' (lower) answers to six questions relating to five computer engineering courses Further, 61.7% of the students marked that they "deal with solving open-ended small-scale problems or design tasks to meet given specifications and constraints," which are mid-level assignments in the taxonomy; only 31.1% agreed that they "are engaged in challenging projects"; 44.2% marked that they "face difficulties"; only 3.3% answered that they "work in teams;" and 26.2% answered 'Very much' or 'Much' for the expression "we develop creativity." To summarize the teachers' answers, we calculated the percentage of teachers who marked 'Very much' or 'Much' for the six items for each of the five course categories. The outcomes are presented in Fig. 1. It may be seen that 100% of the teachers marked 'Very much' or 'Much' for the item "The students deal with basic exercises, drills and practice," which are basic assignments in the taxonomy (scale) we presented earlier in this document; 92.3% marked that the students "deal with solving open-ended small-scale problems or design tasks to meet given specifications and constraints," which are mid-level assignments in the taxonomy; however 0% think that the students "are engaged in challenging projects"; 61.5% marked that the students "face difficulties"; 30.7% answered that the students "work in teams;" and 76.9% answered 'Very much' or 'Much' for the expression "the students develop creativity." These results indicate that in learning computer engineering, the students deal merely with doing basic exercises and solving simple problems. Much work is required to shift the teaching and learning of embedded engineering and computer science towards enhancing students' higher-order cognitive skills such as problem solving and creativity, and fostering teamwork in the engineering class. These are among the major objectives of the E2LP project. IV.
CONCEPT OF THE UNIFIED PLATFORM
The E2LP project concerns a novel development of a unified learning platform for embedded system design, which would serve as a general educational framework for future embedded system engineers. E2LP is supporting the following learning objectives: (1)
embedded microprocessors & computer architectures programming (software aspects),
(2)
digital signal processing (audio, video and data) and its real-time implementation,
(3)
FPGA digital system design and verification,
(4)
FPGA accelerated computing,
(5)
networks & interfaces,
(6)
system integration.
A. Base Board with Extension Boards In its essence, the E2LP platform consists of the Base Board and a set of extensions boards. The E2LP base board is supporting learning objectives (3) and (4). In order to support other learning objectives, at least 2 extension boards will be designed:
microcontroller board, based on ARM-v7 with low power RF IC; supports learning objective (1), (4) and (5),
DSP board based on Marvell ARMADA 1500; supports learning objective (2).
The E2LP base board and these 2 extension boards are together supporting learning objective (6). They will be explained in more detail in section V. The E2LP Learning Platform provides an advanced hardware platform that consists of a low cost Spartan-6 Platform FPGA surrounded by a comprehensive collection of peripheral components that can be used to create a complex embedded system. Additionally, software IDE is developed to support usage of the board.
B. Basic Set of Exercises As embedded designs are becoming more complex, reconfigurable technology is now being seen as a viable option to speed up the embedded engineering design process. This evolution has not only increased the performance of the technology but also put before us the challenge of teaching this technology to computer engineering graduate students. Since this technology is very dynamic, it is essential to teach students the latest design methodologies, based on unified platform using FPGA technologies with advanced sets of exercise. The basic set of exercises [27] presents library of laboratory examples developed for the E2LP embedded engineering learning platform. It is part of the E2LP startup kit containing identified topics (objectives) in embedded systems: (1) digital system design, (2) computer system design, (3) digital signal and data processing (2D and 3D), (4) computer network and communications and (5) system integration (system software, android development). Library of exercises represents also some solved examples, removing entrance barrier into targeted learning objectives. Entrance barrier is often considered as a limiting factor for enabling students’ creativity in learning because they may be afraid of unknown. Exercise logistics provides the methodological procedure for creation and implementation of application oriented exercise for embedded systems. It is an educational roadmap for practice for students and it is used for logistics for development of the optimal technical and systematic approach for embedded system application creativity. At this point of the project we have specified E2LP startup kit describing laboratory examples for each learning objective. The intention is to have three categories of laboratory exercises: (1) basic exercise, (2) problem solving and (3) project solution, which will demand more learning effort through challenging exercises. A starting list of laboratory examples is collected online, by using online library assembling questionnaire, distributed through partners within the consortium, including all necessary information about exercises. Complete set of laboratory exercises will define topics in digital design and for each exercise it will classify: (1) learning target in the area of the embedded system learning objectives, (2) theoretical background knowledge necessary to understand particular exercise and (3) instructions how to run it on the E2LP platform. Implementation of the Augmented Reality Interface will facilitate all these items for each particular exercise. E2LP start-up kit will be available online. The developed set of laboratory exercises on E2LP platform will be included in the existing curriculum through academic partners within the consortium for the success evaluation purposes. Identified and recommended updates of the E2LP startup kit will be added to the existing exercise database. It is expected that E2LP courseware will have at least 60 open source laboratory exercises. The main goals of the courseware are (1) to teach students the fundamental concepts in FPGA-based embedded system design within the preselected topics and (2) to illustrate clearly the way in which advanced FPGA-based embedded systems are designed today, using advanced unified platform and design methodologies and tools. C. Augmented Reality Interface The Augmented Reality Interface (ARI) is the tool that visualizes the main characteristics and the invisible principles happening inside the electronic components of the E2LP board. To be efficiently used by students as a learning resource the following requisites have been found:
Students must be presented with a friendly and intuitive tool that they can manipulate without previous knowledge,
It has to be robust and easy to handle, giving students complete freedom to examine the board from any angle,
It must provide the students with all the required information to understand the board and fulfill the exercises: specification of components, steps to perform an exercise, explanations about solutions etc.,
All the information provided has to be displayed on top of the physical board in order that students can associate the physical device with the corresponding functioning and explanations,
It has to use playful elements and minimal dialogue to make this tool appealing to the students and foster creativity and collaboration between them.
Based on the detected requirements and specifications, a first prototype has been built. The main structure is shown in Fig. 2.
Figure 2. ARI general perspective (A) and bottom view (B). The main components of the ARI are:
Articulated arm - Based on the arms of electronic magnifying glasses, it is provided with sensors in each juncture that will send to the mini-pc their position and rotation in real time. The arm will be manipulated by the students to place the touchscreen on top of the electronic board.
Touchscreen - Situated in the head of the articulated arm, the touchscreen will serve, on the one hand, to display the video stream coming from the webcam and, on the other hand, make selections in the different menus that will appear on it.
Webcam - The webcam, placed in the bottom-center of the touchscreen will be in charge of sending the latter the pictures captured.
Mini-PC - The core of the prototype, the mini-PC will be in charge of the tracking processing and augmented reality visualization.
The interaction between these elements is presented in Fig. 3.
Figure 3. Functional diagram of the ARI. Augmented Reality consists on the superposition in real time of virtual information over the vision of the real world taken from a camera. It usually relies on a marker-based tracking system, marker-less tracking system or a geo-location system to determine whether and where to display the virtual information (Fig. 4). The first two options are based on software processing and depending on the quality of picture given by the camera, processing power of the used device (computer, phone tablet etc.) and other factors like lighting conditions or occlusions of the makers/images, the results can be limited in processing speed and performance.
Figure 4. Marker-based, marker-less and geo-location-based tracking systems. The prototype presented in this paper will introduce hardware tracking information to the image-based tracking software to improve the overall stability, efficiency and performance. Marker-less or image-based tracking consists of the detection and tracking of objects of the real life, like an electronic board. It is performed by searching characteristic points or features of the images, using in this case corner-based feature detector algorithms. For the prototype it has been decided to use OpenCV [28], an open source computer vision library, as it has all the necessary capacities for the picture processing and tracking. The detection of characteristic points is carried out by the ORB (Oriented BRIEF) algorithm, which offers the detection of partially rotated pictures. Fig. 5 shows the final result of the process, where after using different algorithms, the correct key points are detected. The colored picture corresponds to the webcam input and the grey picture is the stored image to be found in the video (the ORB algorithm works on grayscale images).
Image from the webcam Image previously stored in the program.
Figure 5. Matching of keypoints. As it has been described in the previous section, the prototype consists on an articulated arm that can be manipulated by students to situate the camera and screen to different heights and angles regarding the board. To cover the different heights at which the camera can be placed regarding the position of the board, it has been working with 21 different pictures of the board, divided as shown in Fig. 6. Thus, the developed software has to compare the image coming from the webcam with the all the 21 pictures one by one to determine which case is the most suitable one to track.
FAR DISTANCE: The whole board is in view.
MEDIUM DISTANCE: The image of the board is divided in four quadrants that are processed separately.
NEAR DISTANCE: For a close look to the components of the board the image is divided in sixteen quadrants and they are processed separately.
Figure 6. Pictures to be processed according to the distance of the camera regarding the board. Initially, the distance between the camera and the image is calculated by software, but the amount of time invested in this process can be drastically reduced making use of the hardware tracking information extracted from the sensors of the articulated arm. Placing sensors in each articulation of the arm and situating the system at a specific distance of the board, the geometric model of the arm will allow the system to know the position of the center of the camera regarding the board.
z x A
y
B
Figure 7. The (x,y,z) components to extract (A) and the geometric model of the arm (B). The z value of this point (height) will determine the set of pictures to be chosen in Fig. 7, and the (x, y) values will determine the exact image of that set the camera is looking at. Therefore, the software will only have to compare the frames coming from the webcam with the chosen image and superpose the corresponding augmented information to it. This combination of software-based and hardware-based tracking will not only add speed to the system but it will also make it more robust as, on the one hand, the possible occlusions or light changes produced by the manipulation of the system from the students will be compensated by the hardware tracking and, on the other hand, the precision of the sensors of the arm will be improved by the software tracking. At this moment the development of the prototype has been done under Linux OS as it gives the best performance augmented reality. However, there is an open line researched in parallel under Android OS as it is a fast growing technology that in a near future can become the predominant one, with its extended use in smartphones and tablets. The long-term objective of this project is not only that students will use the augmented reality capabilities within the laboratory, but that they will be able to make use of some of those capacities directly on their smartphones, either if they are in classroom or at home.
Once the board is detected and its position tracked, the system must display on top of it the augmented information. This information will differ according to the needs students have when using the ARI, and its use has been classified according the three different kind of exercises defined in this project: basic exercises (BE), problems (PB) and projects (PJ). (Table 1 on the next page) For the displaying system it has been chosen OpenGL [29], the open graphics library that allows the creation of the augmented reality layer, displaying the required information over the video stream layer, and that will be used due to its standardization, multiplatform nature and the fact that it provides a version for embedded systems (OpenGL|ES) that could be used in an Android OS System if in the future the Android developments allow a better performance of the augmented reality than Linux based systems. With the prototype mounted, the following steps are its testing with the E2LP board and its validation in classrooms. Before using it in an educational environment different tests will be made to determine the tracking precision, the speed, the robustness against occlusions and the usability of the ARI. D. Remote Laboratory Laboratories, which are found in all engineering and science programs, are an essential part of the education experience. Not only do laboratories demonstrate course concepts and ideas, but they also bring the course theory into alive. In a traditional laboratory, the user interacts directly with the equipment by performing physical actions (e.g. manipulating with the hands, pressing buttons, turning knobs) and receiving sensory feedback (visual and audio). However, equipping a laboratory is a major expense and its maintenance can be difficult. [30] Since the experiments are performed in a laboratory that contains expensive equipment, the students must be supervised which limits the time they have. This also requires a class with many groups performing the experiment at the same time, and thus many instruments are required to support each group. Laboratory experiments are also a serious problem for distance learning students who may not have an access to the laboratory at all. [31] In E2LP project a Remote Laboratory (RL) is an experiment, demonstration, or process running locally to design and control a RL equipment and software as well as an experiment board based on an FPGA device, but with the ability to be monitored and controlled over the Internet (future E-learning portal).
USE
DESCRIPTION
GENERIC
The Generic Use allows to access information about the board without having to complete any exercise. It will be available at any time as the information it provides is considered basic for the understanding of the composition of the board.
Used with BE
PB
PJ
X
X
X
X
X
X
Names and datasheets. Students will be able to access at any time the specifications of the components they want to study.
The Basic Use of the ARI makes reference to the least information a user needs to take up an exercise: 1) Wording of the exercise. BASIC 2) Prerequisites: - Theory to understand the exercise (concepts introduction). - Links to other exercises related to the current one.
3) Needed HW. If specific hardware of the board is going to be used for an exercise, the ARI will highlight that hardware on top of the board.
An Advanced Use of the ARI will allow users to receive more complex information superposed to the board:
ADVANCED
1) Steps or Instructions to follow: If exercises can be divided in smaller subproblems with a list of steps to be carried out or instructions to be followed, the ARI will show an animation of those steps on the board to help the students understand what they have to do.
X
X?
2) Results and explanations: Basic exercises and some of the problems will have pre-known solutions that can be shown to the students so they know if they have done it correctly. Besides, the ARI can represent an explanation of why/how the solution is achieved, using visual animations through the screen. Table 1: Uses of the ARI according to students’ needs and exercise types.
In the base case, the RL can be an experiment board connected to a computer through a standard interface and with the host computer connected to the Internet and has to provide remote access. The client can be any computer connected to the Internet with an ability to see the same interface as the local host and also have the same programs, interfaces, modules etc. E2LP RL should allow the user to do several actions over an Internet connection, which are the list of E2LP Remote Laboratory functionalities: 1. Dedicated software and hardware solutions will provide an access to laboratory equipment and enable students to set them up and operate them at the required level to carry out exercises. 2. Users could remotely program given set of exercises provided by Start-up Kit over the Internet and observe in near real time the E2LP board behavior. 3. Users could access the essential data sheets and software tools. Tutorials and essential data sheets will be provided and available on the E-learning portal as an introduction to the course. 4. Users could program the experiment base board and simultaneously could monitor the evolution of the experiment via camera interface. What makes E2LP platform innovative is that when students interact with the GUIs, they are actually operating real instruments that are set up in a laboratory in some remote location, controlling them over the internet. The output they see on the instrument's display panels is not a simulation, but is actual data being read from the real instruments, in real time.
Figure 8. Remote Laboratory concept of solution The Fig. 8 above presents remotely controlled environment concept of solution. The whole environment is managed by powerful E2LP Server (WWW/SQL/FTP/RDP/RTSP services), which is equipped with all common interfaces, which are essential for internal hardware compatibility. E2LP Server is connected via Ethernet interface to the local network, which is responsible for seamless data communication between environment’s components. The crucial component of the remotely controlled environment is an experiment base board, which is controlled by programming device (Xilinx Platform HW-USB-II-G). This programming device provides integrated firmware to deliver high-performance, reliable and user-friendly configuration of the base board and enables user to program other Xilinx CPLD devices. This programming device is fully integrated and optimized for use with specialized Xilinx iMPACT software, which enable users to perform remote operations such as programming and configuring FPGA. In remote operations user run iMPACT on one computer but the operations are performed on a device attached to another computer through a Xilinx Cable Server. Xilinx ISE includes such program as well as provide a set of programming tools, which allow user to perform operations remotely. To use this functionality user only needs to specify a remote server address in proper configuration in iMPACT software. This is the most important feature of programming device, from the RL point of view. One of the most important functionalities of the RL is a possibility to control/monitor an experiment base board over local network. To achieve this, it is necessary to forward data directly to the server over common interfaces (like programming device above connected with USB) or over local network by using dedicated hardware solutions and specified proper router configuration. Connection with the RL is provided via e-learning portal, which is also an RL’s content management system (based on Moodle Platform) and is based on Apache server, PHP and SQL server. It provides an access to knowledge (exercises, data sheets) through a web user interface and has an ability to exchange information between laboratory hardware and software applications. The second role of e-learning portal is management of user that is enable them access to the laboratory hardware and software (booking functionality). The E2LP remote laboratory could be equipped with additional real laboratory equipment like Oscilloscope, Digital Multimeter, Function Generator or similar LabView virtual machines in accordance to laboratory exercises necessities. E. Evaluation Methodology It is widely agreed that a major objective of engineering education in the 21st century is fostering students' higher-order thinking skills, such as the ability to analyze and interpret data, identify, formulate, and solve complex engineering problems, and think creatively. However, engineering educators also recognize that developing these skills can take place only if students also acquire basic knowledge and skills in their specialization area such as mechanical, electronic or computer engineering. In addition, engineering education has been increasingly influenced by learning theories such as constructivism and situated learning [32]-[33], according to which human beings construct new knowledge through activity and interaction with the real world, as well as through collaborative and social interaction. Therefore, the most important question in designing and evaluating an educational program is what are students' activities, rather than what instructors do. With these thoughts in mind, we developed the evaluation methodology for E2LP based on the three-level task taxonomy (Fig. 9). We distinguish between:
Exercises: basic closed-ended tasks in which the solution is known in advance and students can check their answers.
Problems: open-ended small-scale tasks in which students might use different solution methods or arrive at different solutions.
Projects: broad challenging tasks in which the problem is ill-defined. For example, a design task in which the learner should determine the objects, identify the constraints and choose the solution method.
The task taxonomy described above, which is derived from the literature on problem solving in engineering [34]-[35], is intended to help in developing instructional materials for teachers and students using the E2LP platform, as well in evaluating the program implementation and outcomes.
Pro jec ts Probl ems Exercises Figure 9. Tasks taxonomy in engineering education
USB
USB
13
29
User I/O
Ethernet
Ethernet
Xilinx FMC LPC
18
4
RS232
Snapwire Connector
4
RS232
74 IR recv
Infra Red
Line In\Out
Audio Codec HDMI TX
VGA
VGA
1
12
28
Power Supply
Xilinx Spartan-6
1
30
CVBS out
Video Encoder
32
Video Decoder
15
7
47
8
DDR2
MMC
CVBS in
FLASH
Figure 10. E2LP Base Board
ARM COP
USB
V.
BOARDS FOR THE UNIFIED PLATFORM
This section will present in more detail the base board and two extension boards which are initial hardware components of the learning platform. A. Base Board The E2LP Base Board performs the following functions:
based on FPGA, provides the central point of the E2LP platform on which all other parts are connected;
supplies power for the whole E2LP platform;
controls programming the FPGA and CPUs on extension boards;
provides a basic user interface;
provides storage, multimedia and communication interfaces for the platform;
provides the platform for digital system design;
provides test points for debugging.
The key building modules of the E2LP Base Board (Fig. 10) are:
Xilinx Spartan-6 FPGA,
ARM-based control processor,
Mezzanine connector to extension board (Xilinx FMC LPC standard),
DDR2, flash and multimedia card memory,
user interface (8 switches, 6 buttons, 8 LEDs, alphanumeric LCD screen),
snapwire connector,
CVBS video encoder and decoder,
video output (VGA, HDMI),
audio sub-system,
communication interfaces (USB, Ethernet, RS232 and Infra-red).
The E2LP Base Board, as presented in this document, together with its extension boards, is working in fully satisfying the main requirement of the E2LP platform – to be used in the complete embedded engineering curriculum and significantly reduce the overhead in engineering education. Implementation of the extension boards whose mechanical requirements are dependent on the Base Board will be explained in the next section. B. Extension Board with ARMADA The extension board based on Marvell ARMADA 1500 [36] has the following functions:
based on ARM processor, provides the extension to the E2LP platform suitable for highly sophisticated signal processing and execution of real-time software,
connects to the E2LP base board via Mezzanine connection,
connects to the exterior with USB, LAN and HDMI interfaces,
provides the extension to the E2LP platform suitable for implementing laboratory exercises in the field of digital signal processing, real-time system software, computer networks and system integration,
provides test points for debugging,
The block diagram (Fig. 11) gives a high level overview of the E2LP extension board based on Marvell ARMADA 1500 processor.
Xilinx FMC LPC
Marvell ARMADA 1500
USB
USB
Ethernet
Ethernet
FLASH
HDMI RX
HDMI TX
DDR3
Figure 11. E2LP Extension board with ARMADA
C. Extension Board with NXP Basic ARM microcontroller extension board has the following components:
LPC2364 microcontroller [37],
High-Precision 1-Wire Digital Thermometer,
Low Voltage Audio Power Amplifier,
Digital accelerometer and I2C,
Snap-wire connector with 8 pins,
Push-button switches, rotary encoder and LEDs,
High speed CAN transceiver.
Extension boards presented in this document allow achievement of education goals in areas in which the Base Board alone could not be used – signal processing, computer networks, system software and application software. The extension board based on Marvel ARMADA 1500 will support courses in signal processing, computer networks, system software, application development and wireless communications. The second extension board, based on a simpler NXP ARM-7 will support courses in the first and second year of engineering education, i.e. computer architecture and basic programming courses. VI.
INITIAL PLATFORM EVALUATION
The following are the requirements with the platform satisfies:
Platform hardware allows students to experimentally verify their abstract design by providing them with the FPGA on which the digital systems and computer system designs can be verified and, by the means of extension boards, providing them with the processor on which they can verify their software algorithms.
Library of laboratory exercises for the platform contains a detailed documentation which explains the problem of the exercise, provides an overview of the required theoretical knowledge and leads the student to a solution without revealing the actual steps and decisions a student must make.
Platform supports education in the board range of topics required for a successful education of an embedded computer engineer, which is a central goal for E2LP project: o o o o o o
digital system design (supported by E2LP Base Board with FPGA), computer system design (supported by E2LP Base Board with FPGA), computer architectures (supported by NXP ARM extension board), digital signal processing (supported by Marvell ARMADA extension board), computer networks & interfaces (supported by Marvell ARMADA extension board), system integration (supported by FPGA and ARMADA).
The project completed its first year and the main tasks were to design the platform and make the initial set of laboratory exercises. At the same time prototype for ARI was designed and initial e-learning portal established. Therefore, the main evaluation steps in the first year were to verify that the platform technically supports the requirements and that the boards operate properly. Evaluation of the educational quality of the platform, which is going to be the main evaluation of the project, will be performed in the next two years. Platform will be used in all courses of the 2nd and 3rd year of the Department of Computing and Systems Control on the Faculty of Technical Sciences, University of Novi Sad, which are in the field of Computer Engineering and Communications. It will also be used in several courses on the Faculty of Electrical Engineering and Computing, University of Zagreb. In-class usage will be a new way of verifying the platform during the project’s 2nd year. VII. CONCLUSIONS This paper presented an Embedded Computer Engineering Learning Platform which aims to be used in the complete curriculum and reduce the overhead in engineering learning. It will ensure a sufficient number of educated future engineers in Europe, capable of designing complex systems and maintaining a leadership in the area of embedded systems, thereby ensuring that our strongholds in automotive, avionics, industrial automation, mobile communications, telecoms and medical systems are able to develop. In such a manner, the E2LP intends to increase European competitiveness in the learning process of embedded computer engineering, ensuring further technological and methodological development of the educational approach in this field. In the next year, platform will be verified in action, by being used by students in the laboratory classes. Since live usage always brings unnoticed vulnerabilities to the surface, platform will be updated to fix these, should they emerge. Platform will be extended with the augmented reality interface and remote lab, to make it a complete system for education of embedded computer engineers. ACKNOWLEDGMENT The research leading to these results has received funding from the European Union’s Seventh Framework Programme (FP7/2007-2013) under grant agreement no 317882.
[1] [2]
[3]
[4] [5] [6] [7] [8] [9] [10] [11]
[12] [13] [14]
REFERENCES K.E. Holbert, G.G. Karady: “Strategies, Challenges and Prospects for Active Learning in the Computer-Based Classroom”, IEEE Transactions on Education, Volume 52, Issue 1, Feb. 2009 pp. 31 – 38 R. Schneiderman: "Economy and Shortages Affect the European Job Outlook", IEEE Spectrum, March 2010, Available from (accessed on 2011-12-10) http://spectrum.ieee.org/at-work/tech-careers/economy-and-shortages-affect-the-european-joboutlook/0 The European Centre for the Development of Vocational Training (Cedefop): "Skills supply and demand in Europe, Medium-term forecast up to 2020", Publications Office of the European Union, 2010, ISBN 978-92-896-0536-6, Available from (accessed on 2011-12-10) http://www.cedefop.europa.eu/en/Files/3052_en.pdf Bureau of Labor Statistics, U.S. Department of Labor, Occupational Outlook Handbook, 2010-11 Edition, Available from (accessed on 2011-12-10) http://www.bls.gov/oco/ M.A. Thompson: "Employment Outlook Asia: Focus on China and India", 2011-04-20, Available from (accessed on 201112-10) http://www.goinglobal.com/articles/864/ S. Nooshabadi, J. Garside: “Modernization of teaching in embedded systems design-an international collaborative project”, IEEE Transactions on Education, Volume 49, Issue 2, May 2006 pp. 254-262 A. Bindal, S. Mann, B.N. Ahmed, L.A. Raimundo: “An undergraduate system-on-chip (SoC) course for computer engineering students”, IEEE Transactions on Education, Volume 48, Issue 2, pp. 279-289 R.B. Foist, C.S. Grecu, A. Ivanov, R. Turner: “An FPGA Design Project: Creating a PowerPC Subsystem Plus User Logic”, IEEE Transactions on Education, Volume 51, Issue 3, pp. 312-318 J.O. Hamblen: “Using a Low-Cost SoC Computer and a Commercial RTOS in an Embedded Systems Design Course”, IEEE Transactions on Education, Volume 51, Issue 3, pp. 356-363 I. Velez, J.F. Sevillano: “A Course to Train Digital Hardware Designers for Industry”, IEEE Transactions on Education, Volume: 50, Issue: 3, pp. 236 – 243 The Joint Task Force on Computing Curricula: IEEE Computer Society & Association for Computing Machinery: Curriculum Guidelines for Undergraduate Degree Programs in Computer Engineering, IEEE Computer Society, 2004, Available from (accessed on 2011-08-11) http://www.acm.org/education/education/curric_vols/CE-Final-Report.pdf K.G. Ricks, D.J. Jackson, W.A. Stapleton: "An Embedded Systems Curriculum Based on the IEEE/ACM Model Curriculum", IEEE Transactions on Education, vol.51, no.2, May 2008, pp.262-270 T.S. Hall, D.V. Anderson: "A Framework for Teaching Real-Time Digital Signal Processing With Field- Programmable Gate Arrays", IEEE Transactions on Education, vol.48, no.3, pp. 551- 558, Aug. 2005 D.T. Rover, R.A. Mercado, Z. Zhang, M.C. Shelley, D.S. Helvick: "Reflections on Teaching and Learning in an Advanced Undergraduate Course in Embedded Systems", IEEE Transactions on Education, vol.51, no.3, pp.400-412, Aug. 2008
[15]
[16] [17]
[18]
[19]
[20]
[21] [22] [23] [24]
[25]
[26] [27]
[28] [29] [30] [31] [32] [33] [34]
[35] [36] [37]
J.W. Pierre, F.K. Tuffner, J.R. Anderson, D.L. Whitman, A.H.M.S. Ula, R.F. Kubichek, C.H.G. Wright, S.F. Barrett, J.J. Cupal, J.C. Hamann: "A One-Credit Hands-On Introductory Course in Electrical and Computer Engineering Using a Variety of Topic Modules", IEEE Transactions on Education, vol.52, no.2, pp.263-272, May 2009 M. Ravel, M. McDermott: “An Electronic System Design Platform for SYSTEMatic Learning in ECE and ICT Curriculum”, 2007 International. Conference on Microsystems Education (MSE’07), pp. 145 – 146 M. Ravel, M. Chang, M. McDermott, M. Morrow, N. Teslic, M. Katona, J. Bapat: “A Cross-Curriculum Open Design Platform Approach to Electronic and Computing Systems Education”, IEEE International Conference on Microelectronic Systems Education, 2009. MSE apos;09., 25-27 July 2009, pp.69 – 72 D. Majstorovic, E. Neborovski, M. Katona: “A Cross-Curriculum Embedded Engineering Learning Platform”, 33rd International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), 2010, Vol. IV. pp. 283-288 I. Kastelan, M. Katona, D. Majstorovic, E. Neborovski: “Students Perspective of the Embedded Engineering Learning Platform - a Case Study in Digital Design”, 34th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), 2011, Vol. IV. pp 215-219 I. Kastelan, D. Majstorovic, M. Nikolic, J. Eremic, M. Katona: “Laboratory exercises for embedded engineering learning platform”, 35th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), 2012, pp 1113-1117 R.T. Azuma: A survey of augmented reality, in Teleoperators and Virtual Environments, 1997, Vol. 6, pp. 355–385 R.T. Azuma, Y. Baillot, R. Behringer, S. Feiner, S. Julier, B. MacIntyre: Recent advances in augmented reality, IEEE Computer Graphics and Applications, 2001, Vol. 21, pp. 34–47 P. Milgram, F. Kishino: A taxonomy of mixed reality visual displays, IEICE Transactions on Information Systems, 1994, Vol. 12, pp. 1321-1329 H. Lukasser, H. Beer, T. Meister, R. Peters: Production scenario of EADS military aircraft of the ARVIKA project – cable loom assembly. Demonstration within the framework of the ARVIKA project 2002. Available at : http://www.ismar2002.org/demos/ismar_lukasser.pdf U. Neumann, A. Majoros: “Cognitive, performance, and systems issues for augmented reality applications in manufacturing and maintenance”, IEEE Virtual Reality Annual International Symposium, Atlanta, Georgia, United States, 14-18 March, 1998. ACM Press, New York, pp. 4–11. C. Baber: “Wearable computers: a human factors review”, International Journal of Human-Computer Interaction, 2001, Vol. 13, pp. 123–145. M. Temerinac, I. Kastelan, K. Skala, B. Medved Rogina, L. Reindl, F. Souvestre, M. Anastassova, R. Szewczyk, J. Piwinski, J.R. Lopez Benito, E. Artetxe Gonzalez, N. Teslic, V. Sruk, M. Barak: “E2LP: A Unified Embedded Engineering Learning Platform”, 16th Euromicro Conference on Digital System Design (DSD), 2013, in press OpenCV open source library, available at http://opencv.org/ OpenGL open source library, available at http://www.opengl.org/ Distance-Learning Remote Laboratories using LabVIEW; www.ni.com A. Nafalski, J. Machotka, Z. Nedic: “Collaborative Remote Laboratory NetLab for Experiments in Electrical Engineering”, Using Remote Labs in Education. Two Little Ducks in Remote Experimentation, 2011, University of Deusto, pp. 177-199 J. Piaget, To Understand is to Invent. New York, Grossman, 1973. L. S. Brown, A. Collins, S. Duguid: “Situated cognition and the culture of learning”, Educational Researcher, 1989, 18(1), pp. 32-42. H. L. Plants, R. K. Dean, J. T. Sears, W. S. Venable: “A taxonomy of roblem-solving activities and its implications for teaching”, in J. L. Lubkin, (Ed.), The Teaching of Elementary Problem-Solving in Engineering and Related Fields. Washington DC, American Society for Engineering Education, 1980, pp. 21-34. M. Barak: “Teaching engineering and technology: cognitive, knowledge and problem-solving taxonomies”, Journal of Engineering, Design and Technology, 2013, pp. 316-333 Marvell ARMADA 1500 http://www.marvell.com/digital-entertainment/armada-1500/ NXP LPC2364 microcontroller http://www.nxp.com/documents/data_sheet/LPC2364_65_66_67_68.pdf BIOGRAPHIES
Ivan Kastelan, born in 1985, received the BSc and MSc degrees in electrical and computer engineering from the Faculty of Technical Sciences, University of Novi Sad, Serbia in 2008 and 2009 respectively. He received the award for the best student of the Faculty of Technical Sciences in 2008. He is currently pursuing the PhD degree and working as a teaching assistant in the Computer Engineering and Communications Department. He is also a research assistant in RT-RK Institute for Computer Based Systems. His current research interests include digital system design and algorithms. During the PhD studies he worked on the hardware and algorithms for automated verification of digital television sets and touchscreen-based devices. He is currently a coordinator deputy and work package leader in the EU FP7 project “E2LP – Embedded Engineering Learning Platform”.
Jorge R. Lopez Benito, IT and Audiovisual Specialist, working 15 years as a System Analyst and System Administrator expert, in the fields of Internet, Cloud and Virtualization for large public and private enterprises. With studies in Computer Engineering from the University of Deusto, currently is the CEO of CreativiTIC Innova, a micro-SME ICT engineering start-up that offers solutions in the crossroads of Augmented Reality (AR), innovative audiovisual technologies and cloud computing.
Enara Artetxe Gonzalez received her Bachelor's Degree in Telecommunications specialized in Telematics in the University of the Basque Country. She is co-founder and Chief Creative and Development Officer in CreativiTIC. She is in charge of investigating new technologies in Augmented Reality and their possible implantation in the products and services of the company, development of augmented reality applications and 3D modeling for those applications.
Prof. Moshe Barak is currently the Head of the Department of Science and Technology Education at BenGurion University of the Negev. He received his PhD degree in Education in Science and Technology from the Technion – Israel Institute of Technology (1986), where he served as lecturer and senior research associate. His research interests focus on fostering higher-order cognitive skills such as problem solving and creativity in science and technology education. Prof. Barak has over 30 years of experience in curriculum development and evaluation in areas such as control systems, robotics, image processing and electronics, including teachers’ pre-service and in-service training. He has been engaged in a range of studies on the cognitive and affective effects of project-based learning and using information and communication technologies (ICT) in science and technology education. Jan Piwinski, MSc. Working in PIAP since 2008. Research specialist, has experience in analysis and development of system applications for security systems. In recent years he has been involved in systems development, the HMI design and optimization as well as techniques of information and knowledge presentation. Prof. Miodrag Temerinac received the Ph.D. degree in electronics and computer engineering from University of Belgrade, former Yugoslavia, in 1983. He is Alexander-von Humboldt fellow and he spent two years (1988-1990) having Humboldt fellowship at the University of Hanover in Germany. In 1992 he changed to industry joining Micronas GmbH in Freiburg, Germany working on the IC development and later switching to management. In 2005-2006 he spent one year as the director of system development in the new founded R&D Center in Shanghai. His fields of interest are DSP algorithms and architectures, hardware/software co-design of complex systems on chip and product development in consumer electronics. He is a coordinator in the EU FP7 project “E2LP – Embedded Engineering Learning Platform”.
