Augmented reality (AR) is a technology that shows promise for creating specific and highly customized engineering applications. Probably the best example that encapsulates these qualities is the innovative AR startup DAQRI, who just released SmartGlasses at CES 2017. Its Smart Helmet was designed to be American National Standards Institute certified (it is in the process of gaining certification), and it builds custom applications for industrial engineering sectors.

Microsoft HoloLens, which was just recommended by GE at its last Mind + Machines event, is probably the most well-known AR device out on the market. Niantic Incorporated deserves a mention for popularizing smartphone AR with its hit Pokémon Go in 2016, which helped lift the term “augmented reality” to a sturdier place in the popular vernacular.

But teaching engineering students isn't about popularity. It’s about helping students understand how best to apply science and mathematics when tackling real-world problems. Sometimes, there’s an answer to be had from well-worn techniques, but there will always be tough problems, and inventing and innovating new technological solutions is the only way to break through.

What is the best way for today's engineering students to expand their spatial acuity and geometric awareness? Perhaps these students can learn from education programs made by computer-aided design (CAD) vendors like SOLIDWORKS?  After all, the importance of creating and transferring ideas from 2D to 3D cannot be understated.

But who better to help engineer a better approach to teach students than an engineering professor?

One such person is Professor Modris Dobelis of Riga Technical University (RTU) in Latvia, who has been developing ways to teach engineering students to improve their spatial reasoning using AR. When it comes to understanding, interpreting and analyzing digital drawings of engineering structures, it’s easy to understand how the immersive quality of AR technology could be tempting to try out as a promising new teaching tool.

In a research paper coauthored by Dobelis and Veronika Strozheva entitled “Application of Augmented Reality for Teaching Descriptive Geometry and Engineering Graphics Course to First-Year Students” in 2014, principal author Zoja Veide describes their desire to understand how AR could positively impact the ability of engineering students to visualize and manipulate geometrical shapes. Because the program was designed for first-year students, the researchers could collect data that would allow them to see how well AR impacts the ability of students to improve their spatial comprehension and graphic representation skills from the beginning of their higher education.

The primary reason these researchers were curious about the ability of AR to help engineering students was time. Lecture hours were limited at RTU in Latvia for students engaged with engineering curricula that deals with engineering graphics. The researchers found that they were presenting condensed versions of lessons about spatial and graphic representation to students who were taking a course called “Descriptive Geometry and Engineering Graphics.”

To give students taking the class a chance at a more rewarding and better outcome, the teachers turned to something called the AR-DEHAES toolkit, a work of software that was developed by a group of professors in Spain. In a paper entitled “AR-DEHAES: An Educational Toolkit Based on Augmented Reality Technology for Learning Engineering Graphics,” authors Jorge Martin-Gutierrez, Jose Luis Saorín, Manuel Contero, and Mariano Alcañiz Raya outline the origin and creation of the software application.

AR-DEHAES contains 100 exercises with virtual 3D models that are visualized on a 2D screen. The exercises contained in the “augmented notebook” provide fiducial markers of virtual 3D models that students can manipulate to see different perspectives and complete each exercise.

Another reason that AR is attractive to teachers as a potential teaching tool is because students must pay attention to boring engineering graphics courses, which have a high withdrawal rate. 

However, the skills taught in these AR and engineering graphics classes are critical for future engineers who are continuing their journey to graduation and developing their skills and habits with CAD engineering software. One way these classes can learn from AR is the value of getting a student's hands dirty. Perhaps getting student's hands on the CAD software, even as a trial bases, could be beneficial.

How Is Professor Dobelis Using Augmented Reality to Teach Engineers Now?

Dobelis is currently teaching a course called “Parametric Modeling of Machine Elements.” The main objective students are expected to accomplish is to be able to pass the Certified SOLIDWORKS Associate test.

At the end of the course, students receive an assignment to reverse engineer a machine of seven to 10 parts. Students draw the assembly in SOLIDWORKS and create an animation, rendering or a simple finite element analysis. At the end of the course, the students are expected to have gained an understanding of what makes up a general engineering mindset and understand some procedures based around an improved ability with an engineering software.

Interview With Professor Dobelis

How did you come to use the AR-DEHAES toolkit to teach engineering students?

Primarily, we use the AR-DEHAES toolkit just for the introductory course “Descriptive Geometry and Engineering Graphics.” In 2011, my colleagues met the developers from the University of La Laguna in Spain in one conference, and they shared a free toolkit and assignments for us to test at RTU. We translated the Spanish materials into Latvian, and in the fall of 2011, we put them on the RTU internal e-learning portal, which is based on Moodle.

It was supposed to foster the development of spatial imagination for engineering students to compensate for the lack of preliminary drafting practice, knowledge that was mastered in high schools in Latvia more than 20 years ago. We thought the AR “toy” would attract or entertain students to spend more time training to increase their spatial abilities.

The AR support materials in the form of additional exercises were implemented and tested in one group, and the experiment was done by Zoja Veide. A survey finalized the course outcomes, and there are some publications regarding this experiment. Another experiment was performed to prepare AR solutions as assistant models for solving conventional 2D engineering drafting problems. We developed our own 3D models of simple parts with SOLIDWORKS and then converted them into AR models with freeware for students to use in classic course. The third experiment was done in the course “Computer Graphics in Civil Engineering.”

What do you hope to accomplish with AR in engineering education?

The topic “roof and drawings” involved a classic assignment using a 2D graphics pencil and/or AutoCAD and a 3D model with ArchiCAD. As an additional assignment for students, it was suggested to produce an AR solution from an ArchiCAD model or make a development for a cardboard roof model. Some of these experiments are described in some publications in which the first author and main researcher is Zoja Veide.

Do students enjoy you AR engineering graphics class?

Yes, of course. The survey confirmed that they did enjoy to play. But the design and production of these toys is time consuming and not a rewarded enterprise at RTU.

What are some challenges you face when designing and programming AR for engineers?

Challenges can present unexpected drawbacks. One serious local drawback was the enormous additional work required to check the created assignments manually. Due to overload, we simply could not afford to spend more time on checking additional submissions because the staff’s work with practical and lab assignments is underestimated. It is more profitable to lecture in large groups of students and use computerized tests rather than work in small groups and frequently assess a lot of manual drawings. Another obstacle we faced last year was the inability to buy any AR software we wanted—even cheap ones. The general problem in Latvia is critically low financing for university education. And even when we had the budget for the software, we failed to get it during the entirety of 2016 because of bureaucracy issues — such as organizing procurement procedures. In other words, the academic freedom of universities in Latvia is handcuffed by bureaucracy. So, for now, all the research is mainly restricted by the available freeware.

How important is the AR-DEHAES toolkit to developing your curriculum?

The AR-DEHAES solutions for the course “Descriptive Geometry and Engineering Graphics” used at our department before are frozen for now and not used. At present, we keep working on the developing 3D models for introductory engineering graphics courses with the hope that we’ll get them later to use in AR materials when things improve. We just repeated what other professionals, particularly from the University of La Laguna, accomplished long ago. Respect to them!

How did SOLIDWORKS first come to RTU?

Advancement with SOLIDWORKS teaching activities in the courses of RTU was possible thanks to the European social funding project “The Harmonization of CAD Communication Skills for Concurrent Engineering.” During the planning of the project, from 2005-2008, a new computer lab with 30 powerful CAD workstations, SOLIDWORKS and other CAD licenses was acquired. At our department of computer-aided engineering graphics, SOLIDWORKS is taught in the following courses: “Machine Element Parametric Modeling,” “Computer Graphics in Engineering Communications”, “Fundamentals of Graphics Communication,” and “Computer Aided Design.” It is also suggested as “toy” for self-studies for those advanced students who are bored in the classic PAD (pencil-aided design) course “Descriptive Geometry and Engineering Graphics.” The total number of students in these courses who get acquainted with SOLIDWORKS every academic year is about 50. However, this is not the only one department that teaches SOLIDWORKS at RTU.

What are some of the assignments you give to students and how do you decide how best to leverage SOLIDWORKS?

In our department, we basically use our own study materials in Latvian as they are developed and adjusted for the local student performance and mentality and adapted for the allowance of course credit. Typically, the students have some compulsory classes and homework that cover the basics of part, assembly and drawing modes and a semester-long project involving a reverse-engineering project. For the semester project, they are asked to choose any kind of existing design consisting from seven to 10 parts that could be disassembled or orthographic drawings of any project they could find with/without dimensions. The task is to create models of parts and assembly in SOLIDWORKS using a bottom-up design approach; however, in a very few cases, we challenge them to try a top-down modeling approach. The idea of the reverse-engineering project is to encourage the students to explore SOLIDWORKS' potentials beyond the limits of what is covered in the class. When the students have to solve a practical problem, they are more interested and the end result is better than when dealing with abstract theory. Some of them got so excited and addicted to SOLIDWORKS that they mad assemblies with more than 100 parts. The following figures demonstrate the complexity of the best reverse-engineering projects.

The student responses to questionnaires at the end of the course show that we managed to prove the importance of classic descriptive geometry and engineering graphics knowledge in the contemporary product lifecycle management technologies by means of CAD software.

Coming up with new ways to keep students engaged while guiding them through material that might seem dry or boring is challenging. What is one of the best techniques you've used with students that had the most positive results?

Our goal was to use it as an attractive and motivating “toy” to engage the students in their studies. The student responses to a question about what module of SOLIDWORKS they enjoyed most of all and what they would like to study in more detail never resulted in an answer. Therefore, it is not a surprise that our employers in accreditation boards started to complain about the decreasing ability of graduates to interpret or understand engineering drawings (blueprints). To contribute to an interest in comprehension of engineering drawings beyond the course, we organized competitions for motivated students who wanted to prove their knowledge. This annual event is called the Olympic CAD and PAD contest because it was performed in two tracks—CAD and PAD. The type of tasks varies from year to year and is always different for each track. Usually two astronomic hours are given for the participants to prove their knowledge.

One type of task for the CAD track was checking their ability to “read and write” an assembly drawing of some mechanical device consisting of six to 11 components. SOLIDWORKS was used as a “writing” tool to prove their understanding of how to prepare exact models of components, the form and size of which must be determined from the assembly drawing with just a few given references and overall dimensions and annotations. Another type of task was preparing an exact model of rather complex mechanical parts represented in an orthographic detail drawing with all dimensions and annotations provided. No pictorial representation was given, but after the contest, the students were acquainted with the real part they modeled just few minutes ago, and they could instantly judge what they missed while interpreting the Multiview drawing. An example of an orthographic drawing of the part in Figure 1 shows the complexity used in the fall 2015 semester.

The last tasks were prepared both in Latvian and English so that the students from the foreign student department studying in English also could participate. We also issued certificates acknowledging participation and handed out small prizes. One year, a local SOLIDWORKS reseller, PLM Group Latvia, sponsored the event and handed out wristband USB pen drives for the top six contestants.

In 2011, the assembly drawing “reading and writing” idea was developed further, and I prepared a project called, “Evaluating Engineering Graphics Literacy in the CAD Age,” which I used for an application for a Fulbright scholarship program.

The aim of the proposed research project was to elaborate, suggest and test in the practice one of the methods for engineering graphics literacy quantitative evaluation – ability to “read and write” engineering documents using CAD media.

It is a well-known fact that engineering graphics education in the CAD age is still an important issue. During the last three decades, computers and IT applications have changed the engineering education drastically. Introduction of different computer science subjects has turned some graphic literacy developing subjects into a second-class category. The result is that the time allotted for teaching engineering drawing in many universities has been reduced two to four times compared to the pre-computer age.

The lessening of content and the partial withdrawal or underestimation of the importance of some core subject knowledge like descriptive geometry and engineering graphics occurred all over the world, starting with technology developed countries and following the others. Already about two decades ago, it was pointed out that this is not the right way to go.

A Fulbright scholarship was granted, and the project was carried out in the spring 2011 semester at North Carolina State University (NCSU) in Raleigh, N.C., in cooperation with Ted Branoff from the College of Education of Department of Science, Technology, Engineering and Mathematics.

The developed “reading and writing” test involved assessing the ability to “read” or understand a multiview assembly drawing of a mechanical device that consists of six to 11 parts. Understanding should be proved by representing these parts in engineering language either by “writing” them in an engineering drawing in a conventional way on paper, or digital media using a 2D drafting technique or creating a geometric model of the part using 3D modeling software.

It is also a well-known fact is that, using constraint-based 3D solid modeling software, it is possible to express one’s understanding of visual form much faster than creating a multiview working drawing. The test is highly oriented on the spatial reasoning of simple geometric forms that are present in the parts of the multiview assembly drawing rather than checking the CAD software usage skills.

To complete this test, only the basic knowledge of the 3D modeling technique is required. This allows the students to focus more on the main task of how to express interpreted information and how to build the models from simple 3D geometric primitives like a prism, cylinder, cone, sphere, helix and so on.

Ten mechanical devices consisting of six to 11 parts with different levels of complexity were selected and modeled with SOLIDWORKS to check the model consistency. A wide range of typical elements like threads, chamfers, fillets, notches, grooves, and slots were present in the parts of these devices. Several devices also included springs.

Multiview assembly drawings with parts list were printed and used for practical training and pilot test purposes. Minor editing was applied to the drawings printed in PDF format since it was required to represent threaded joints in sectional views per the conventional standards. An example of the assembly is given in Figure 2, and the following appendix shows the table characterizing the complexity.

This approach ensured that the assembly drawings prepared this way will be most suitable for the “reading and writing” test, and they could not be easily misinterpreted because they weren’t clear enough. Both first- and third-angle projection multiview assembly drawings of B or A3 sizes were prepared.

Educators from several universities worldwide were addressed and suggested to perform this “reading and writing” test using parametric feature-based modeling software like SOLIDWORKS, Solid Edge and Inventor. Alternatively, they were allowed to organize the test in a modified way even with 2D drafting software and submitting their rules and results for comparison.

A positive response was received from two universities in Latvia, one in Croatia and one in the Russian Federation, and they joined the research to test the suggested methodology conducted from NCSU. The staff from these universities received both sets of drawings for training and the actual test. However, not all the universities used the SOLIDWORKS 3D modeling approach.

From the Fulbright project report—complexity and pictorial view of parts from the VALVE PNEUMODEVICE assembly.

A suggested quantitative evaluation method for interpreting the mechanical assembly drawing via the “reading and writing” test in combination with 3D parametric modeling could be used for accurate measuring of the graphic literacy level of the engineering students. One of the main drawbacks is a concern about the ability to scale up the project to handle more students because the time required assessing the student files is very high.

More detailed information about the results of this research could be found in publications, e.g. Google search by keywords “Dobelis Branoff graphic literacy.” In some publications, it is mentioned as the Riga-Raleigh “reading and writing” test, referring to the places it was developed and approbated. Original training and test sets in a PDF format are available for researchers or educators if anybody wants to repeat the Riga-Raleigh test. What was previously only available only in large and rich companies was made accessible to every student.

Participation in this project provided a great experience for a further career in research and education. The set also includes an assessment rubric in a Microsoft Excel spreadsheet with an explanation and sample of evaluated student files.

How did this affect the success rate of engineering students looking to receive SOLIDWORKS certification?

So, the students at RTU also took a chance to pass the Certified SOLIDWORKS Associate academic certification exam. The success rate for the fall 2016 semester was six out of 11 students from our engineering courses. This is an excellent tool for automated assessment of the students’ experience and practice and is highly oriented towards SOLIDWORKS usage skills. From my opinion, the modeling of parts and assemblies given in the pictorial views is much easier than dealing with real-life engineering orthographic or multiview detail drawings.

The last experiment performed during the fall 2016 semester was testing the recently developed computer-assisted assessment tool GraderWorks. We wanted to figure out whether or not it could be successfully used to automate the process of assessing students’ SOLIDWORKS parts and assembly model files during the semester.