Adaptive Tracking in Mobile Robots With Input-Output Linearization
- 9 July 2014
- journal article
- Published by ASME International in Journal of Dynamic Systems, Measurement, and Control
- Vol. 136 (5), 054503
- https://doi.org/10.1115/1.4027369
Abstract
This paper presents a neural network adaptive controller for trajectory tracking of nonholonomic mobile robots. By defining a point to follow (look-ahead control), the path-following problem is solved with input-output linearization. A computed torque plus derivative (PD) controller and a dynamic inversion neural network controller are responsible for reducing tracking error and adapting to unmodeled external perturbations. The adaptive controller is implemented through a hidden layer feed-forward neural network, with weights updated in real time. The stability of the whole system is analyzed using Lyapunov theory, and control errors are proven to be bounded. Simulation results demonstrate the good performance of the proposed controller for trajectory tracking under external perturbations.Keywords
This publication has 4 references indexed in Scilit:
- Robust adaptive control for a nonholonomic mobile robot with unknown parametersControl Theory and Technology, 2009
- EP-based kinematic control and adaptive fuzzy sliding-mode dynamic control for wheeled mobile robotsInformation Sciences, 2009
- Designing a robust adaptive dynamic controller for nonholonomic mobile robots under modeling uncertainty and disturbancesMechatronics, 2003
- Control of a nonholomic mobile robot: Backstepping kinematics into dynamicsJournal of Robotic Systems, 1997