Introduction

The development of new embedded algorithms and control techniques for dynamic systems such as motors, industrial processes, automobiles, and aircraft, amongst others, normally requires real-time tests in the devices to be controlled. However, such tests using real systems can be very expensive and time-consuming. For example, the refinement of a control algorithm for an automobile injection valve requires tests under real conditions that can last for hours or even days. In other cases, it is necessary to construct small prototypes in order to perform the tests, as in the case of aircraft and ships. Many of these tests under real conditions can be performed using off-line simulations (which are not performed in real time), although such simulations do not fully eliminate real-time testing.

Methods for the development of platforms for real-time simulation have been extensively studied for many years. However, research in this area has recently accelerated following advances in terms of speed and the ease of development associated with new hardware platforms. Methodologies for real-time simulation have included the use of hardware such as digital signal processors, general-purpose processors, and even reconfigurable computational solutions employing field-programmable gate arrays (FPGAs). The benefits of using real-time simulation platforms have been demonstrated in various areas of graduate teaching. In an approach different to those of the works mentioned above, this project presents a platform for distributed real-time simulation, employing a variety of hardware platforms in order to provide greater modularity and robustness in the simulation process.

The main objective of the Real-Time Simulator for Dynamic Systems (RTSDS) is to provide a support tool for the developers of embedded control systems, focusing on reducing the costs involved in real-time tests. In addition to having industrial locations, the RTSDS could also be used in the academic sphere as a programmable didactic model in various curriculum modules of courses in Engineering. Using the RTSDS, the graduate or postgraduate student could test control algorithms in various types of dynamic system, simply and at a cost that is relatively low compared to those of real didactic models. Another point that should be emphasized about the proposal presented here is that the RTSDS can also simulate the functioning of the drivers and sensors associated with the control of dynamic systems, which is an important characteristic from the point of view of the development of control systems projects.