K. Varun Raj, Sharschchandra V. Bidargaddi, K.N. Krishnanand, D. Ghose
ABSTRACT: This paper presents VITAR (VIsion based Tracked Autonomous Robot), a robotic test bed that consists of a tracked mobile robot equipped with a pan-tilt mounted vision system, an onboard PC, driver electronics, and a wireless link to a remote PC. A novel vision-based obstacle avoidance algorithm is implemented on the robot. The robot uses histograms of images obtained from a monocular and monochrome camera to detect and avoid obstacles in a dynamic environment. As the generation of histogram is computationally inexpensive, the robot is quick to avoid the obstacles. Several experiments conducted with different obstacle environments validate the effectiveness of the algorithm.
KEYWORDS: Tracked Mobile Robot, Vision-based Obstacle Avoidance, Image Segmentation
full paper (pdf)