What remote labs can do for you

The Internet of Things (IoT) is the network of physical devices connected to the internet. Online connections enable users to remotely monitor the devices and their surroundings or to actively control them through sensors and actuators.

As the technology has progressed, the importance of the IoT has grown tremendously. According to the McKinsey Global Institute, the IoT’s annual economic impact could reach $11 trillion by 2025. That figure represents about 10% of today’s world economy. ¹ Bullish sounding at first, the prediction is not so ridiculous if one stops to think about how deeply internet technologies have already penetrated today’s society.

Consider, for example, mobile devices—those smartphones and tablets everybody carries nowadays. A typical modern smartphone has numerous sensors that allow it to capture the device’s orientation, location, ambient light conditions, and much more. And it is frequently connected to the internet. A world filled with such internet-connected devices opens new and exciting possibilities in industry, economy, research, and many other sectors, not least among them education.

Labs at a distance

Remote laboratories (RLs), which can be used for industrial or research purposes, ² connect lab equipment to the internet so that they can be operated at a distance. For example, many telescopes and instruments at large particle-accelerator-based facilities can be remotely monitored and, at least to some extent, operated. RLs are one of the most direct applications of the IoT to education, especially in physics and other fields for which experimentation is a key part of the learning process. Teachers and students typically access an RL through a Web application, ^{3 , 4} through which they can operate the laboratory equipment, perform an experiment, and obtain real data.

Depending on whether it uses commercial or open-source technologies and what type of experiment it performs, an RL can be either exceedingly expensive or amazingly cheap to implement. A remotely operated radio telescope is obviously more expensive than an RL setup to validate Archimedes’s principle. The same considerations also apply to how complex or simple an RL can be and how much work is required to maintain it.

Today RLs are an important topic for the education community. ^{5 , 6} International conferences are held every year to discuss the impact, importance, benefits, and limitations of using multimedia and technology in science, technology, engineering, and mathematics education. Relevant examples of such conferences are Multimedia in Physics Teaching and Learning, Remote Engineering and Virtual Instrumentation, Experiment@ International, and Internet Based Control Education. Attendees from around the world learn about and present new advances and applications.

The importance of students having physical access to laboratory exercises is not in doubt. However, anecdotal evidence and formal studies indicate that RLs have at least two important advantages. ^{3 , 7} First, unlike in traditional hands-on labs, students in RLs can complete or repeat experiments on their own time, and they have greater flexibility to conduct the experiments from any location. Second, RLs can provide access to a range of equipment that may be too expensive, dangerous, or logistically problematic for a particular high school or university.

Through RLs, universities and other educational institutions can offer their students more experiments and more experimentation time, and they can do so in a cost-effective and flexible manner. The case is especially true for students with physical disabilities who may have limited access to laboratories, and students in high schools or distance universities, which do not hold classes at a physical location. And because the labs can be accessed 24 hours a day, 7 days a week, the instrumentation remains idle less often.

Experiments on radioactivity or ones that use high-power lasers are common examples of lab activities that might be more appropriate for students to perform remotely. An additional advantage of RLs is that they can be combined with augmented reality to highlight an effect or phenomenon that is not normally visible to the naked eye (see figure 1). Some studies indicate that virtual (simulated), remote, and hands-on labs can equally enhance student understanding of concepts when compared with instruction that has no practical component. ^8–10

Although many studies show that virtual labs are as effective as RLs for helping students learn and understand concepts, two factors make RLs more interesting. First, the real data that RLs produce come with experimental uncertainty—something students must become accustomed to facing. Second, in today’s research environment, scientists often control equipment remotely, and the practice will likely become even more widespread in the future. Students must be prepared to work with RLs and understand the problems that may arise when doing so.

Remote complications

Although the experiments are performed remotely, RLs still require supervision from staff members. The degree of supervision depends on the complexity of the experiment. Moreover, safety issues may exist for some setups. For example, RLs that use high-voltage or high-current applications may require the physical presence of staff to mitigate fire risks during experimentation. In addition, technological limitations preclude RL implementation of many experiments in chemistry, biology, and other subjects. Although RLs provide experiences that hands-on labs cannot, the opposite is also true, at least for now.

The most frequently expressed complaints about RLs are that they don’t sufficiently develop students’ experimental skills, they don’t sufficiently hold students accountable for the data they generate, and they encourage students to think in a menu-driven way. A further concern is that they rob students of the opportunity to screw things up and thereby gain insights into instrument limitations and observational techniques.

For many of the problems facing RLs, especially those related to implementation difficulties, it is only a matter of time before they are solved or at least mitigated. Other concerns, such as the risk of students thinking in a menu-driven way or not having the opportunity to make mistakes, require different approaches or methodologies when developing an RL and the interface used to operate it. For example, virtual reality could be used to avoid menu interactions. Imagine a three-dimensional virtual world in which, through an avatar, you grab, move, and plug in instruments and components. Also, a carefully designed RL can allow some errors to happen during an experiment. Thus some complaints mentioned previously should be better understood as criticisms of particular RL implementations rather than of the RL concept itself.

Remote labs in practice

Regardless of their advantages and disadvantages, every year an increasing number of universities begin to use RLs. A decade ago they were only available at the few universities with at least one research group that specialized in the technologies. Today many universities offer RLs as a complement to traditional hands-on laboratories, with two objectives in mind: Give students the ability to work with lab experiments they otherwise could not do, and give them more lab time.

The University of Amsterdam, for example, provides such access through four RLs (www.nat.vu.nl/~gerritk ). The labs are aimed at senior high school students and first- and second-year undergraduates. They offer open-inquiry-based investigations; they are designed so that students define their own questions and then use the labs’ applications to conduct experiments in search of answers. A more in-depth data and error analysis is required from university students, so their RL user interface contains more advanced tools. High school students get a simpler interface and less freedom to adjust experimental settings.

In the Faculty of Applied Informatics of the Tomas Bata University in Zlín, the Czech Republic, all physics labs are performed using experimental setups with data acquisition cards connected to computers. In the lab, students use computer software to collect and plot the data from the experiment. The same setups are offered remotely (www.ises.info/index.php/en/laboratory ) so that students can repeat or complete the experiments on their own time and from their homes.

The Remote Farm (http://remote.physik.tu-berlin.de/en ) from the Faculty of Mathematics and Natural Sciences of the Technical University of Berlin offers 15 physics RLs. The first one went on line in 2004. An interesting feature of the remote farm is that it provides open, live video streams of all its RLs. Although only one registered user at a time can access a given experiment, anyone can watch how an experiment is carried out when that particular experiment is in use.

The previous three cases are examples of universities and groups that use RLs to complement and enhance the labs they offer to their own students or those at a limited number of affiliates. In the past several years, however, as the interest in RLs and the economic support for them have increased, larger projects involving multiple institutions have appeared. ¹⁰ Those networks were born of the idea to create a formal context in which multiple institutions could easily group and share their RLs. Thus a wider educational community could benefit from their use. Students that access an RL in one of the networks may perform an experiment without knowing where the setup is located.

The sharing of experiments enables participating institutions to offer a more extensive catalog. Such is the case of iLabCentral (http://ilabcentral.org ), the Labshare Institute (www.labshare.edu.au ), and Graasp (http://graasp.eu ), three international efforts that share RLs in an open way. The networks are likely to become a good entry point, if they are not already, for physics teachers who want to start using remote labs as a complement to their lessons.

Good RLs are especially in demand for distance education. The Open University of the UK and the National Distance Education University (UNED) of Spain are distance universities whose students are scattered throughout their respective countries. Students attending the universities often juggle jobs with their studies and have tight restrictions on their schedules.

Offering laboratory activities as part of the curriculum at a distance university is not trivial. More than a decade ago UNED turned to RLs so the students could perform experiments wherever they are. To learn about one of UNED’s RLs, turn to the box below. A noteworthy aspect of how UNED uses RLs is that students are always encouraged to first perform an experiment with a virtual lab before doing it again with an RL. The students can then compare the theoretical results, given by a mathematical model, with real results. The practice is, in fact, one of the fundamental steps of the scientific method. In recent years additional Spanish universities have joined with UNED to share RLs in a network called UNILabs (http://unilabs.dia.uned.es ).

In 2012 the Open University established the OpenScience Laboratory, an initiative to give teachers a large number of experimental resources, among them two robotically controlled telescopes: a radio telescope in Milton Keynes, UK and an optical telescope in Mallorca, Spain, (see figure 2). The telescopes are available online. However, due to high demand and the enormous complexity of the astronomical devices, access to the facilities is restricted to those who undergo appropriate training. (The Open University is willing to share the telescopes. Details on how interested educational institutions can apply for access to the telescopes can be found on the OpenScience website, https://learn5.open.ac.uk ). The OpenScience Laboratory is growing rapidly and plans to add several new RLs to its webpage in the near future.

What the future holds

To date, all RLs have been ad hoc stand-alone solutions with limited or no capability to cooperate with other platforms. ¹¹ In particular, the way the computer applications are programmed to connect to hardware, how those applications are deployed on the Web (through a learning management system, for example), and the way the RL is shared among institutions (if at all) depend on the developer of each particular lab. Therefore, the RL community has begun to discuss ways to standardize those aspects.

Standardized communication protocols for connecting with the hardware devices used by RLs would mean a complete decoupling of the lab application view (client side) and the experiment (server side). That decoupling would enable multiple lab interfaces to control an RL without the need to modify anything in the experimental setup or the program that controls the devices. Those different interfaces could be designed with different student backgrounds in mind. Moreover, whereas the lab applications currently need to be created by a programmer, a standard would open the door for the automatic generation of such applications. The P1876 working group of the IEEE Standards Association is already making efforts to define a standard for those protocols.

Standardized Web applications for RLs would enable developers to deploy and connect the applications easily with a variety of content and learning management systems, among them WordPress, Drupal, Moodle, and Blackboard Learn. Applications that comply with the standard could communicate with those Web environments so that teachers, managers, and administrators could track how much, when, and how they are used and grade activities related to the laboratories. The applications could also allow users to save data files or images captured during an experiment and store them in the cloud.

Automating the way an RL is shared is even more complicated. On the one hand, the lab owners must be able to use their own equipment whenever they need it. On the other hand, if an RL is to be successfully shared, those who want to use it must also be able to count on good service in terms of scheduling and accessibility. Currently RLs are shared mainly as a result of agreements made between the lab owners and lab consumers. In the future, we can expect automated sharing of RLs to be not so different from, and not much more difficult than, the way people today share simulations, videos, and manuscripts.