Rough Terrain Autonomous Vehicle Control Using Google Cloud Vision API

Abstract

Computer vision (CV) represents a technology using which information contained in a visual scenery is obtained by capturing and digital processing of the images. Computer vision is becoming more popular and widely used in automated vehicles, robotics to emulate human vision capability and industrial automation. Google cloud vision API is a remote real time image processing platform to process, classify and acquire required metadata from image and videos. In this work, design and implementation of a rough terrain autonomous vehicle using Rocker Bogie mechanism and Google cloud vision API for obstacle avoidance is presented. Rocker Bogie mechanism is employed for maneuvering the robot on rough terrains.