Obstacle detection & elimination of shadows for an image processing based automated vehicle

Abstract

In today's world, Artificial Intelligence (AI) has begun to take center stage. The recent trend has been moving towards the creation of robotic devices which are capable of making spontaneous and effective decisions in demanding situations. One of the challenges faced in such a scenario is dealing with the permutations and combinations of countless real-life circumstances, most of which require to comprehensively train the machine so that it is capable of taking appropriate actions, all the while taking care of any exceptions. Human errors may cause disastrous effects in certain circumstances; such errors can be reduced using AI. One of the most important uses of AI in recent times has been its application in automated vehicles. If the movement of vehicles were to be automated, much time can be saved, and moreover, some amount of travel related stress for humans is lifted. In order to implement the automation of locomotives, detection of any obstructions in the vehicle's path is a must. Also, the algorithm should be robust enough to detect any type of obstacle irrespective of it being a vehicle, barrier or people crossing the road. A successful attempt to perform the same has been discussed in this paper, through a novel yet simple algorithm utilizing image processing techniques. The simulation of the algorithm was performed using the Simulink tool in MATLAB, version 2012a. A success rate of 93% was achieved with respect to the detection of obstacles.