IRIS: A Wearable Haptic Jacket for Practical Navigation for the Visually Impaired People

Abstract

Around 2.2 billion people worldwide live with some form of vision impairment. Traditional white canes cannot reliably detect obstacles above waist height. This paper presents IRIS, an open-source prototype of a wearable haptic jacket that gives vibration cues about nearby obstacles. The system has two parts. The first uses three ultrasonic sensors (left, center, right) connected to an ESP32 microcontroller to measure obstacle distance, and a linear mapping converts each distance into the intensity of the vibration motor on the same side. The second uses a Raspberry Pi camera that sends an image to the Azure Computer Vision service every three seconds; detected objects are assigned a direction and an approximate proximity, which trigger the corresponding motors, and urgent hazards are signaled with repeated high-intensity vibration bursts on the same motors. The two devices communicate over Wi-Fi using UDP. The prototype is built from off-the-shelf components, and its code and firmware are released under the MIT license. The camera part depends on an internet connection and a cloud service.