Design and evaluation on adaptive fuzzy speed control of mobile robot

Wheeled mobile robots are widely used in various fields such as agriculture, industry, land mining, military, space explorations, and other applications in which the environment is inaccessible or hazardous to human, such as in nuclear plants. This thesis mainly focuses on investigation of fuzzy logic approach capability in order to control the speed of a four-wheel mobile robot in an indoor environment. This research deals with adaptive fuzzy speed control of mobile robot in an indoor environment with variable slope. Two different method are used for tuning mechanism consist of on-line tuning of output gain and output membership functions. Given a reference trajectory, the performances of these methods are compared with fuzzy PI controller (FPIC), through experimental evaluations. The first method is self-tuning fuzzy logic controller by means of updating output scaling factor. Depending on the process trend, the output scaling factor (SF) of the controller is modified by an updating factor (α), in an online fashion. The value of α is determined through a rule-based adaptive mechanism defined over error and the pitch angle of the robot. In the second adaptive method, membership functions tuning, a direct adaptive fuzzy controller is used to modify the locations of the output membership functions, adaptively. The controller has a fuzzy rule base that can adopt different output membership functions for each fuzzy rule to improve the performance of the fuzzy controller, in spite of changes in the plant. The effectiveness of the control system is verified through real time experiments in an indoor environment with different slope. Both data acquisition and control algorithm are developed by using LabVIEW. A four-wheel mobile robot, PUTRABOT2, is used to conduct the experiments. Performance comparison between the fuzzy PI controller and adaptive fuzzy controllers are made in terms of several performance criteria including rise time, settling time, peak time, peak overshoot, integral absolute error (IAE) and integral time weighted absolute error (ITAE). Comparative results for various processes in term of different angle of ramp (15, 20 and 25 degree) show that the adaptive fuzzy controllers outperform the fuzzy PI controller to minimize the rise time, settling time IAE and ITAE, except peak overshoot. This research paves the way towards the adaptive control of mobile robots in the face of plants uncertainty and would be appreciated by urban and industrial applications.

Preview PDF FK 2011 73R.pdf Download (965kB) | Preview

Actions (login required)

View Item