Adaptive fuzzy sliding mode and indirect radial-basis-function neural network controller for trajectory tracking control of a car-like robot

Document Type: Research Paper


1 Dept. of Electrical Engineering, Amirkabir University of Technology, Tehran 15875-4413, Iran

2 Dept. of Automotive Engineering, Iran University of Science and Technology, Tehran 16846-13114, Iran

3 Department of Automotive Engineering, Iran University of Science and Technology, Tehran 16846-13114, Iran

4 Department of Electrical Engineering, Sharif University of Technology, Tehran, Iran



The ever-growing use of various vehicles for transportation, on the one hand, and the statistics of
soaring road accidents resulting from human error, on the other hand, reminds us of the necessity to
conduct more extensive research on the design, manufacturing and control of driver-less intelligent
vehicles. For the automatic control of an autonomous vehicle, we need its dynamic model, which,
due to the existing uncertainties, the un-modeled dynamics and the performed simpli cations, is
impossible to determine exactly. Add to this, the external disturbances that exist on the movement
path. In this paper, two adaptive controllers have been proposed for tracking the trajectory of a
car-like robot. The rst controller includes an indirect radial-basis-function neural network whose
parameters are updated online via gradient descent. The second controller is adaptively updated
online by means of fuzzy logic. The proposed controller includes a nonlinear robust section that
uses the sliding mode method and a fuzzy logic section that updates some of the nonlinear control
parameters online. The fuzzy logic system has been designed to deal with the chattering problem in
the controller of car-like robot. In both controllers, the parameters have been determined by means of
genetic algorithm. The obtained results indicate that even with the consideration of un-ideal effects
such as uncertainties and external disturbances, the proposed controller has been able to perform