Abstract:

The acquisition of practical skills and experimentation plays a vital role in technology and engineering education [1]. However, laboratory teaching in Automatic Control and Robotics becomes a challenge for large groups of students (probably with more than 6 people). In that situation, one of the most effective and reasonable alternative to support teaching is to use simulators or interactive tools in order to avoid students to queue to test their designs in a real platform, reducing consequently the wasted time. Furthermore, the use of real platforms needs a heavy budget in terms of resources, especially related to multiplicity of equipment. Due to these facts, teachers and researchers have developed a huge variety of resources to support teaching in laboratory courses. In the Automation field, there is a considerable number of specialized and general purpose modelling tools available for that end (see e.g. [2,3] and references therein for a current review). Among them, in this paper we focus on those included in the MATLAB/Simulink package, mainly due to their extended use for both the design and the simulation of control systems.

Nowadays, the potential of Simulink has grown especially by adding a multitude of toolboxes, among which we are interested in emphasizing Simscape for physical modeling. It is an object-oriented modeling tool which is spreading in current simulation environments thanks to the use of the individual components of the model and their interconnections to define the underlying dynamic equations. Simscape is a powerful software package that extends the Simulink product with tools for modelling and simulation of physical systems, such as those with mechanical, electrical, hydraulic, thermal and pneumatic components. Unlike other Simulink blocks, Simscape ones directly represent physical elements or relationships. Then, a system model can be built in this environment in the same way that a physical system is assembled. Although many example of Simscape-based resources can be found in the literature, there are very little examples to support teaching and learning in classrooms.

Given this context, the aim of this paper is to show the advantages of using physical modelling to develop interactive resources to support teaching and learning in Automatic Control and Robotics. In particular, our interest is to build simulators for laboratory courses with which students can validate, for example, their designed controllers. As illustrative examples, the simulator of the rotatory pendulum Quanser Qube– Servo and a mobile robot will be built in Simscape.
The main contribution of this paper is to present two educational platforms to be used for students in laboratory courses of Automatic Control and Robotics, including educational aspects and experience of use.

References:
[1] Gomes, L., García-Zubia, J. (2007). Advances on remote laboratories and e-learning experiences. University of Deusto, Bilbao, Spain.
[2] Guzmán, J.L., Costa-Castelló, R., Dormido, S., Berenguel, M. (2016). An Interactive-Based Methodology to Support Control Education. IEEE Control Systems Magazine 63, pp. 63–76.
[3] Dormido Bencomo, S. (2004). Control Learning: Present and Future. Annual Reviews in Control 28(1), pp. 115–136.

Keywords:

Technology, teaching, laboratory, course, engineering, simulator, Simscape.