Term projects

In mechanisms with gearbox, a problem of backlash may lead to uncontrollability. This work is to figure out the identification of backlash between teeth of the contacted gears. In this work, identification techniques are proposed based on a contemporary backlash model. An analytic description of this model of backlash with hard dynamic nonlinearity, that depends on proper switching functions, is discussed. The proposed identification techniques are applied to this model to estimate its unknown parameters. These techniques include Iterative Least Square method, Recursive Least Square, Kacz-maz’s Algorithm, Projection Algorithm and Recursive Instrumental Variable method. Simulation studies of all the backlash identification techniques are performed on cascaded systems connected with input backlash. Cases of static and time varying backlash are included.

This work presents the design of a nonlinear control strategy to achieve the swing-up and stabilization of the pendulum bob in a pendulum-cart system. Partial feedback linearization and intuition from the energy dynamics of the system are tools employed in the design of the control functions. Simulation results show the satisfactory performance of the system under the closed loop effect of the designed control.

This work is to figure out Model Predictive control technique on an unmanned quadrotor helicopter. At first we must get the mathematical model of this system and check its linearity. If the system is nonlinear, so we have to convert it to a linear one using a linearization technique. Decoupling technique will be used in deriving the model to study each motion type of the helicopter separately. Then convert the linear model to a state space representation to monitor internal states and control the output using a model predictive control algorithm applied to the state space model. Finally monitor the response of the controller with different parameters and constraints on the system variables to evaluate its performance depending on some pre-described control design criteria using Matlab simulink. Eventually, compare between simulation results to find the optimal parameters.

This work is to figure out the derivation of equation of motion for the rotating spindle of a commercial Hard Disk Drives (HDDs) that exposed to external mechanical vibrations using Lagrange’s Equation. MEMS accelerometer will be used to measure the value of the applied external acceleration, in addition, its effects on the angle of rotation of the spindle are calculated from the derived equation of motion, this effect can be eliminated using a compensator to improve the performance of the HDD. Finally simulating the applied external vibrations with their effects on the spindle angle using MATLAB SIMULINK.

This work is to figure out some control techniques on a 2-DOF Pendulum Driven Spherical Rolling Robot. At first we must get the mathematical model of this system and check its linearity. If the system is nonlinear so, we have to convert it to a linear one using a linearization technique. Decoupling technique will be used in model derivation to study each motion type of the robot separately “rotation about transverse and longitudinal axis”. Then convert the linear model to a state space representation to monitor internal states and control the output using some control algorithms applied to the state space model as, Full state-feedback controller, Integral Control State feedback, Full State feedback controller using LQR optimization technique, PID controller, Model Predictive Control, and Nonlinear control. Finally monitor the response of each controller to evaluate its performance depending on some pre-described control design criteria.

Project brief: The robot is used in factories to lift the product from production line to put it in the storage, in case of facing an obstacle the robot will avoid it then returns back to the original path.

Features:

• Optimum mechanical design which is strong enough to withstand the load.
• Fully automated Mobile Robot.
• Obstacle Avoidance.

Tools and Technologies: Microcontroller PIC16F877A, Fuzzy Logic controller (A.I control), Ultrasonic sensor, Shaft Encoders.

Fuzzy controller: The inputs to fuzzy controller are the distance away from obstacle & motor speed and the output is the change of voltage going to the motors.