Abstract
A person operating a mobile robot in a remote environment receives realistic visual feedback about the condition of the road on which the robot is moving. The categorization of the road condition is necessary to evaluate the conditions for safe and comfortable driving. For this purpose, the mobile robot should be capable of recognizing and classifying the condition of the road surfaces. This paper proposes a method for recognizing the type of road surfaces on the basis of the friction between the mobile robot and the road surfaces. This friction is estimated by a disturbance observer, and a support vector machine is used to classify the surfaces. The support vector machine identifies the type of the road surface using feature vector, which is determined using the arithmetic average and variance derived from the torque values. Further, these feature vectors are mapped onto a higher dimensional space by using a kernel function. The validity of the proposed method is confirmed by experimental results.
Original language | English |
---|---|
Pages (from-to) | 357-363+18 |
Journal | ieej transactions on industry applications |
Volume | 131 |
Issue number | 3 |
DOIs | |
Publication status | Published - 2011 |
Keywords
- Disturbance observer
- Environment recognition
- Mobile robot
- Motion control
- Real-world haptics
- Support vector machine
ASJC Scopus subject areas
- Industrial and Manufacturing Engineering
- Electrical and Electronic Engineering