2021 | OriginalPaper | Chapter
Obstacle Avoidance in Robot Navigation Using Two-Sample T-Test-Based Obstacle Recognition
Published in:
Recent Innovations in Computing
Abstract
In this paper, a solution to the obstacle avoidance problem in autonomous robot navigation is presented using two-sample t-test. A procedure to perform t-test on two-independent LASER scan distance-range vectors is provided. The proposed strategy compares the LASER scan readings of the obstacles in the navigation path. Similarity of the obstacles is recognized on the basis of the acceptance of the null-hypothesis. Robot navigation results are simulated using Turtlebot-Gazebo simulator and MATLAB software. A discussion is made on the results produced using the proposed strategy with the previously existing results in the literature.
