Analysis and implementation for a walking support system for visually impaired people

Visually impaired people are faced with challenges in detecting information about terrain. This paper presents a new walking support system for the blind to navigate without any assistance from others or using a guide cane. In this research, a belt, wearable around the waist, is equipped with four ultrasonic sensors and one sharp infrared sensor. Based on mathematical models, the specifications of the ultrasonic sensors are selected to identify optimum orientation of the sensors for detecting stairs and holes. These sensors are connected to a microcontroller and laptop for analyzing terrain. An algorithm capable of classifying various types of obstacles is developed. After successful tests using laptop, the microcontroller is used for the walking system, named 'Belt for Blind', to navigate their environment. The unit is also equipped with a servo motor and a buzzer to generate outputs that inform the user about the type of obstacle ahead. The device is light, cheap, and consumes less energy. However, this device is limited to standard pace of mobility and cannot differentiate between animate and inanimate obstacles. Further research is recommended to overcome these deficiencies to improve mobility of blind people.