Negative obstacle detection for wearable assistive devices for visually impaired

The objective of this work is to develop a negative obstacle detection algorithm for a robotic wheelchair. Negative obstacles — depressions in the surrounding terrain including descending stairwells, and curb drop-offs — present highly dangerous navigation scenarios because they exhibit wide characteristic variability, are perceptible only at close distances, and are difficult to detect at normal operating speeds. Negative obstacle detection on robotic wheelchairs could greatly increase the safety of the devices. The approach presented in this paper uses measurements from a single-scan laser range-finder and a microprocessor to detect negative obstacles. A real-time algorithm was developed that monitors time-varying changes in the measured distances and functions through the assumption that sharp increases in this monitored value represented a detected negative obstacle. It was found that LiDAR sensors with slight beam divergence and significant error produced impressive obstacle detection accuracy, detecting controlled examples of negative obstacles with 88% accuracy for 6 cm obstacles and above on a robotic development platform and 90% accuracy for 7.5 cm obstacles and above on a robotic wheelchair. The implementation of this algorithm could prevent life-changing injuries to robotic wheelchair users caused by negative obstacles.

Download Full-text

Development and optimization of a portable guidance system for the blind and visually impaired with the application of artificial intelligence

10.31219/osf.io/8ahyf ◽

2021 ◽

Author(s):

Tamas Nemes

Keyword(s):

Artificial Intelligence ◽

Visually Impaired ◽

Obstacle Detection ◽

Guidance System ◽

Performance Standard ◽

Audio Signals ◽

Blind And Visually Impaired ◽

Improve Accuracy ◽

New Type ◽

Map Accuracy

This work describes a new type of portable, self-regulating guidance system, which learns to recognize obstacles with the help of a camera, artificial intelligence, and various sensors and thus warn the wearer through audio signals. For obstacle detection, a MobileNetV2 model with an SSD attachment is used which was trained on a custom dataset. Moreover, the system uses the data of motion and distance sensors to improve accuracy. Experimental results confirm that the system can operate with 74.9% mAP accuracy and a reaction time of 0.15 seconds, meeting the performance standard for modern object detection applications. It will also be presented how those affected commented on the device and how the system could be transformed into a marketable product.

Download Full-text

A Hands-Free Obstacle Detection and Avoidance System

Engineering, Technology & Applied Science Research ◽

10.48084/etasr.562 ◽

2015 ◽

Vol 5 (3) ◽

pp. 801-804

Author(s):

M. Abdul-Niby ◽

M. Alameen ◽

O. Irscheid ◽

M. Baidoun ◽

H. Mourtada

Keyword(s):

Obstacle Avoidance ◽

Visually Impaired ◽

Low Cost ◽

Cost Effective ◽

Obstacle Detection ◽

Text To Speech ◽

Voice Signal ◽

Visually Impaired People ◽

Impaired People ◽

The Voice

In this paper, we present a low cost hands-free detection and avoidance system designed to provide mobility assistance for visually impaired people. An ultrasonic sensor is attached to the jacket of the user and detects the obstacles in front. The information obtained is transferred to the user through audio messages and also by a vibration. The range of the detection is user-defined. A text-to-speech module is employed for the voice signal. The proposed obstacle avoidance device is cost effective, easy to use and easily upgraded.

Download Full-text