Model-based localization for an autonomous mobile robot equipped with sonar sensors

Abstract

The problem of localization of an aoutonomous mobile robot is considered. After proposing a simple geometric representation to account for the vagueness inherent in a sonar reading, interval-valued fuzzy set approaches are utilized within a search mechanism to find non-dominated positions and orientations. The search procedures are tested in two experimental seetings and results are reported.

Adaptive extended Kalman filter (AEKF)-based mobile robot localization using sonar

“…On the basis of the above principles, some tentative methods are used to choose the detailed value of . Q (0) is set to be the initial covariance of the state noise vector, i.e., Q(0) = cov(w(0), w(0)), where w(0) is the initial state noise vector, see equation (9). Generally, the initial three states, i.e., x (0), y (0), and (0), are mutually independent, thus the off-diagonal elements of Q (0) are set to be zero, and the values of diagonal elements are the results of initial errors of x(0), y(0), (0), respectively.…”

“…The orientation deviations estimated using the EKF and the AEKF corresponding to Figure 19. R (0) = cov(v(0), v(0)), where v(0) is the initial observation noise vector, see equation (9). In general, the initial eight sonar observation variables are mutually independent, thus the off-diagonal elements of R (0) are set to be zero, and the values of diagonal elements are the results of initial observation errors of eight sonars respectively.…”

“…The algorithm to achieve this can be classified into two large groups: relocalization through the detection of landmarks present in that environment [4][5][6][7] and relocalization through an a priori map of the environment. [8][9][10][11] Here, we mainly discuss the relocalization algorithm related to the extended Kalman filter (EKF) given an a priori map of the environment. E.T.…”

Interval-valued preference structures

“…The methodology developed in this paper is applied to the localization problem of a mobile robot (Bilgiç 1995, Bilgiç & Türkşen 1995b. In a nutshell, the problem is to localize the robot (finding robot's location) in a given environment by solely depending on its sensors (in this case sonar).…”

“…This is in tune with qualitative analyses which direct one to areas where a pointed quantitative analysis is needed. Details of the search algorithms and the manner interval-valued preference relations are utilized can be found in Bilgiç & Türkşen (1995b).…”