In this paper, we present a system for controlling the angular velocities of the motors of a 2WD mobile robot using an optimal Linear Quadratic Regulator with Tracking (LQRT), thanks to a Co-simulation between two Raspberry Pi modules and the MATLAB R2018a software. Indeed, we have a system made up of a certain number of elements, notably a web interface for communication with users, a Raspberry Pi 4 Model B module that we have configured as a server and a Raspberry Pi 3 Model B module that plays the role of a socket client who’s physical GPIOs are represented in an identical logical manner on Simulink in order to facilitate interaction with the process modelled in MATLAB R2018a. This system has been realized thanks to a combination of various software technologies such as the python flash framework for the development of the web application, the HTML and CSS programming languages for the client side of our user application, the library written in C language SQLite for the relational database engine accessible by the SQL language, the JavaScript library Socket. IO library for real-time bidirectional communication between clients and servers, the Python Threading library to facilitate the execution of parallel processes and the python RPI library to control the GPIO ports of the Raspberry Pi 3. The speed control simulation results on a 2WD mobile robot in both normal and Co-simulation modes show almost identical performance indices.

In this work, we propose a low-cost teleoperable laboratory architecture, in which we integrate a palette of dynamic processes accessible via a mobile platform. In the current health context, our goal is to promote learning in engineering in developing countries via remote laboratories with a low-cost architecture. In this architecture accessible via wifi on a mobile platform, we use an ESP32 microcontroller as master, an ESP-CAM for visual feedback on our processes and several ESP32 microcontrollers as slave depending on the number of processes on our palette. We have chosen the ARDUINO NANO microcontroller to manage each process of the pallet. To test our architecture, we have integrated a process of visualization of the curves of variation of the current and the voltage of the brushless direct current motor according to the speed level. Using experimental methods, the error variations between the values ​​measured with the tachometer and those found by calculation are between 0.00094 and 0.0350. Thus, the PWM setpoint delivered by our architecture is indeed equivalent to the output speed.