Outdoor Obstacle Detection Module to Assist Visually Impaired

The objective of this initiative is to overcome the constraints of current technology used by individuals with vision impairment to improve their autonomous navigation. Current technology such as walking sticks have limitations that inhibit individuals from participating from various activities. To address these restrictions, a hands-free module has been designed. The module enables identification of dynamic objects in outdoor and indoor environments. A Raspberry pi board, a Raspberry pi camera module, a buzzer and ultrasound sensors are some of the components used in this solution. Object detection is being done with the real-time video captured using the camera and the object detection model used has been deployed using TensorFlow Lite. The ultrasonic module simultaneously examines the user’s surroundings in three directions to detect obstacles. The distance between the user and obstacle is calculated with the data obtained from the Ultrasonic sensors. When an obstacle is detected, the user is immediately alerted via the audio output through headphones. This enables them to react to their surroundings swiftly. All of these separate modules are combined onto a belt for convenience and durability which gives a hands-free option to those who need it.