This paper describes a procedure to find the transfer function for the Lego Mindstorms Ev3. Lego Mindstorms Ev3 can serve as the platform for a system modeling and a controller design course. It is economical and accessible. It is also very compatible with Matlab and Simulink. This platform can be used for concepts of modeling, feedback, and controller design. The main approach in this work focuses on the closed loop instead of open loop. Although this approach turns the problem into a more complicated puzzle, it reveals more details. In this work, different techniques have been used, such as time domain, root locus, and least square estimation. Different tools have also been utilized such as Matlab SISO tool, the Matlab System Identification tool, and Simulink. These methods and implementations assisted to acquire different types of transfer functions for the system. By simulating the transfer functions and comparing them with experimental studies, the matching scores were calculated to decide on the best transfer function. Finding the finest transfer function for this gadget enables us to prepare diverse practical undergraduate and graduate curricula.

In this thesis, I present my work on the modeling and control of a motor system using the Lego EV3 robot. The overall goal is to apply introductory systems and controls engineering techniques for estimation and design to a real-world system. First I detail the setup of materials used in this research: the hardware used was the Lego EV3 robot; the software used was the Student 2014 version of Simulink; a wireless network was used to communicate between them using a Netgear WNA1100 wifi dongle. Next I explain the approaches used to model the robot’s motor system: from a description of the basic system components, to data collection through experimentation with a proportionally controlled feedback loop, to parameter estimation (through time-domain specification relationships, Matlab’s curve-fitting toolbox, and a formal least-squares parameter estimation), to the discovery of the effects of frictional disturbance and saturation, and finally to the selection and verification of the final model through comparisons of simulated step responses of the estimated models to the actual time response of the motor system. Next I explore three different types of controllers for use within the motor system: a proportional controller, a lead compensator, and a PID controller. I catalogue the design and performance results – both in simulation and on the real system – of each controller. One controller is then selected to be used within two Controls Systems Engineering final course projects, both involving the robot traveling along a predetermined route. The controller’s performance is analyzed to determine whether it improves upon the accumulation of error in the robot’s position when the projects are executed without control.