ASSISTIVE WALKING STICK FOR VISUALLY IMPAIRED

Abstract

This paper presents the smart walking cane based on ultrasonic sensors, Arduino UNO and concepts of Machine Learning and CNN for visually impaired people. There are approximately 37 million people across the globe who are blind according to the World Health Organization (WHO). People with visual disabilities are often dependent on external assistance which can be provided by Humans, trained dogs, or special aid devices. Thus we are motivated to develop a smart walking stick which not only detects object but also detect the floor, surrounding along with collision detection and raises an alarm in case of any mishap. The system was designed, programmed using C language and tested for accuracy. Our devise can detect objects within 5m of distance