Make your own free website on Tripod.com

Mathematical modeling Simulation of Automatic Control of Wheeled Mobile Robots

 

I worked in this project under Prof. Kurien Issac (I.I.T. Bombay) during my six months practical industrial training. The project is aimed at, making Feedback Control strategies for Automatic control of Wheeled mobile robots. The work includes a detailed study on the analytical study of kinematics of WMR (Wheeled Mobile Robots). The objective is to develop an appropriate kinematic model of a WMR and test various time varying feedback control algorithms on this model to control its motion from a given starting position to desired goal position. Various control strategies are reviewed and compared for trajectory tracking and posture stabilization in an environment free of obstacles. From the comparison of the obtained results, guidelines are provided for WMR end-users. Three different kinematic models (based on two different types of physical models) are developed in the following work and tested using ode23 solver of MATLAB.

The first model is a tricycle type model having two rear wheels driven independently and a front wheel on a castor. The model is tested using a time varying smooth feedback control law satisfying liapunovís criterion for stability. Various modifications in the control strategy are tested and the results are presented. The strategy is then extended so as to make the vehicle trace a number of goal positions.

The second model is similar to a conventional vehicle in which the front wheels can be steered through a range of permitted values of angle in accordance with the longitudinal speed and length of the vehicle. Different control strategies are tested on this model and modified suitably to yield satisfactory results for all situations. This strategy is also extended so as to make the model trace a number of goal positions.

In the third model, the space in the vehicle frame of reference is divided into a number of different geometric regions. The behavior of the vehicle can be modeled in a particular way according to the presence of the vehicle in a particular region of space till it reaches to a very close vicinity of the desired point where it finally stops. A time varying control strategy is tested and the results are discussed. The problems encountered are eliminated by suitable improvements in the strategy.