Affiliation:
1. Northeastern University, China
2. Tokyo University of Agriculture and Technology, Japan
Abstract
Considering the noise effect during the navigation of a two wheeled mobile robot, SVM and LS-SVM based control schemes are discussed under the measured information with uncertainty, and in the different environments. The noise effect is defined as uncertainty in the measured data. One of them focuses on using a potential function and constructing a plane surface for avoiding the local minima in the static environments, where the controller is based on Lyapunov function candidate. Another one addresses to use a potential function and to define a new detouring virtual force for escaping from the local minima in the dynamic environments. Stability of the control system can be guaranteed. However, the motion control of the mobile robot would be affected by the noise effect. The SVM and LS-SVM for function estimation are used for estimating the parameter in the proposed controllers. With the estimated parameter, the noise effect during the navigation of the mobile robot can be reduced.
Reference34 articles.
1. Model selection for the LS-SVM. Application to handwriting recognition
2. A Mobile Robot Path Planning Using Genetic Algorithm in Static Environment
3. Chang, C. C., & Lin, C. J. (2011). LibSVM - A library for support vector machines. Retrieved from http://www.csie.ntu.edu.tw/~cjlin/libsvm/.
4. Practical selection of SVM parameters and noise estimation for SVM regression
5. De Brabanter, K., Karsmakers, P., Ojeda, F., Alzate, C., De Brabanter, J., & Pelckmans, K. … Suykens, J. A. K. (2011). The LS-SVMlab toolbox homepage. Retrieved from http://www.esat.kuleuven.be/sista/lssvmlab/