Tarek A. Tutunji

Mechatronic systems identification using an impulse response recursive algorithm

Abstract A recursive identification algorithm is used to identify mechatronic systems using impulse response data. The algorithm is based on an auto regressive moving average (ARMA) model with a steepest descent method to minimize the least square error between the original and predicted outputs. Two mechatronic systems are tested: DC motor and gyroscope. Impulse voltage input is used to excite the system and the angular speed output is measured. In both systems, the torque and angular velocity outputs are dependent on the voltage and current inputs. This relationship is governed by characteristics such as inductance, resistance, moment of inertia, friction, load, and system constants. Once the ARMA model is constructed, the transfer function is realized. Then the input voltage is varied and the identified model results are compared with the original system. Simulation results using Simulink and experimental results using Labview with data acquisition card (DAQ) are presented. Results show that the recursive identification algorithm is able to identify the two systems with minimal error.

Hardware-In-the-Loop for on-line identification and control of three-phase squirrel cage induction motors

Abstract This paper describes a strategy for identification and control of three-phase squirrel cage induction motors. The strategy in this work is divided into 3 stages: on-line identification, off-line controller design, and on-line control. First, the transfer function is identified on-line. Next, the controller design is performed in a pure simulation environment using the identified transfer function. Finally, the designed controller is applied to the real system. Simulation and experimental results are presented to show the validity of the proposed strategy. Advantages of the proposed strategy include high accuracy in the identified system, simplicity, and low cost.

An Undergraduate Mechatronics Project Class at Philadelphia University, Jordan: Methodology and Experience

Mechatronics is a branch of engineering whose final product should involve mechanical movements controlled by smart electronics. The design and implementation of functional prototypes are an essential learning experience for the students in this field. In this paper, the guidelines for a successful mechatronics project class are presented, evaluated, and discussed. Furthermore, the paper introduces a general mechatronic system design methodology that should equip students to carry out a successful mechatronics project in their undergraduate training. Three student projects at Philadelphia University, Jordan, are examined in detail, with descriptions of their goals, design, and implementation.

Identification and cascade control of servo-pneumatic system using Particle Swarm Optimization

Abstract This paper presents a cascade control methodology for pneumatic systems using Particle Swarm Optimization (PSO). First, experimental data is collected and used to identify the servo-pneumatic system where an Auto-Regressive Moving-Average (ARMA) model is formulated using PSO algorithm. Then, cascaded Proportional–Integral–Derivative (PID) controller with PSO tuning is proposed and implemented on real system using Hardware-In-the-Loop (HIL). The identified model is validated experimentally and the performance of the cascaded-PID controller is tested under various conditions of speed variation. Experimental results show that cascaded-PID with PSO tuning performs better than single-PID, especially in disturbance rejection (a practical challenge in industrial pneumatic systems). Results also show that cascaded-PID with PSO-tuning performs better than cascaded-PID with self-tuning in the transient and steady-state responses.

DC Motor Identification using Impulse Response Data

DC motors are widely used in mechatronic systems. The identification behavior for such motors is of interest. The motor torque and angular velocity outputs are dependant on the voltage and current inputs. This relationship is governed by the motor characteristics such as inductance, resistance, moment of inertia, friction, load, and motor constants. In this paper, impulse response data is used to identify the motor transfer function with voltage as input and angular speed as output. An auto regressive moving average (ARMA) model with a steepest descent algorithm was used to minimize the error between the original and modeled velocities. Once the ARMA model was constructed, the transfer function was realized. Then the input voltage was varied and the identified model results where compared with the original system

Parametric system identification using neural networks

Display Omitted A mathematical relationship, between ANN weights and ARMA parameters, is derived.Transfer functions are approximated from the ANN weights.An algorithm (NN2TF) that approximates transfer functions from ANN models is developed.Simulation runs for time and frequency responses are analyzed.Simulation runs are used to validate the algorithms results. Neural networks are used in many applications such as image recognition, classification, control and system identification. However, the parameters of the identified system are embedded within the neural network architecture and are not identified explicitly. In this paper, a mathematical relationship between the network weights and the transfer function parameters is derived. Furthermore, an easy-to-follow algorithm that can estimate the transfer function models for multi-layer feedforward neural networks is proposed. These estimated models provide an insight into the system dynamics, where information such as time response, frequency response, and pole/zero locations can be calculated and analyzed. In order to validate the suitability and accuracy of the proposed algorithm, four different simulation examples are provided and analyzed for three-layer neural network models.

Fuzzy control of a two-wheel balancing robot using DSPIC

This paper is concerned with the design and implementation of a two-wheel balancing robot. The angle and angle change are used as inputs to the robot system in order to calculate the appropriate motor force to balance the robot. ADXL330 accelerometer and two DC motors are used as the sensor and actuators, respectively. More importantly, the robot system uses fuzzy control that is implemented on DSPIC30F2010. The robot under study represents a mechatronic system because it involves an integrated design among mechanics, electronics, and embedded smart controllers. The system model is first tested under Matlab/Simulink. Then, the Printed Circuit Boards are designed, the C-program is written, and the mechanical structure is built. Finally, the robot is built and tested in the laboratory.

UAV quadrotor implementation: A case study

Experience gained in modifying and implementing a quadrotor, an Unmanned Aerial Vehicle (UAV), is presented. Their design can be divided into three main parts: Mechanical structure, electronics circuitry, and embedded algorithms. Therefore, quadrotors present an excellent platform where students can learn synergetic design of mechatronic systems. This paper provides a practical experience in modifying an existing Remote Control (RC) quadrotor kit for autonomous operations. Specifically, a second microcontroller was programmed and interfaced with the original circuitry kit in order to bypass the RC commands and accept tasks according to functions embedded in the new controller. The importance of the work was the ability to modify an existing hobbyist quadrotor to accept new operations.

A methodology for identification and control of electro-mechanical actuators.

Graphical abstract

Reverse Engineering Course at Philadelphia University in Jordan

Reverse engineering (RE) is the process of testing and analysing a system or a device in order to identify, understand and document its functionality. RE is an efficient tool in industrial benchmarking where competitors’ products are dissected and evaluated for performance and costs. RE can play an important role in the re-configuration and redesign of legacy and/or undocumented systems. It can also play a key role in helping students understand engineering products. This paper presents the Philadelphia University experience in constructing a RE course and integrating it within the engineering curricula. This paper can be used as a guide to other universities that wish to introduce RE to their curricula. The information provided in this paper covers the RE methodology for a system level, as well as mechanical, electronics and software levels. Finally, samples of student projects are presented in order to show the learning capabilities provided throughout the course.