Reliable Ultrasonic Obstacle Recognition for Outdoor Blind Navigation

Abstract

A reliable state-of-the-art obstacle detection algorithm is proposed for a mobile application that will analyze in real time the data received by an external sonar device and decide the need to audibly warn the blind person about near field obstacles. The proposed algorithm can equip an orientation and navigation device that allows the blind person to walk safely autonomously outdoors. The smartphone application and the microelectronic external device will serve as a wearable that will help the safe outdoor navigation and guidance of blind people. The external device will collect information using an ultrasonic sensor and a GPS module. Its main objective is to detect the existence of obstacles in the path of the user and to provide information, through oral instructions, about the distance at which it is located, its size and its potential motion and to advise how it could be avoided. Subsequently, the blind can feel more confident, detecting obstacles via hearing before sensing them with the walking cane, including hazardous obstacles that cannot be sensed at the ground level. Besides presenting the micro-servo-motor ultrasonic obstacle detection algorithm, the paper also presents the external microelectronic device integrating the sonar module, the impulse noise filtering implementation, the power budget of the sonar module and the system evaluation. The presented work is an integral part of a state-of-the-art outdoor blind navigation smartphone application implemented in the MANTO project.

An Extended Usability and UX Evaluation of a Mobile Application for the Navigation of Individuals with Blindness and Visual Impairments Outdoors—An Evaluation Framework Based on Training

“…Nonetheless, throughout this process, we discovered some of the limitations of our current implementation that could potentially be improved. Although we have managed to deliver a functional version of the obstacle detection system [59], there are still scenarios, such as balconies or short trees, that need to be improved in the future. Furthermore, users quite often forgot to activate the public means of transport mode of operation that tracks the route on the bus.…”

“…To better understand the interaction of the aforementioned components, the reader may refer to Figure 1. The MANTO project contribution involving reliable ultrasonic obstacle recognition for outdoor blind navigation is presented in detail in [59].…”

“…However, the servomotor mechanism is indispensable for the configuration of the system utilizing the narrow/pencil-beam sensors due to the smaller required rotation step. For more details regarding the power consumption as well the CPU utilization of the impulse noise filtering software, the reader is advised to refer to [59]. Although a comparison between the power consumption characteristics of the existing hardware solutions would be of great interest, nonetheless, the majority of papers do not include relevant experimentation sections.…”

“…Last but not least, Meliones A. et al [59] proposed an obstacle detection algorithm as a component of a mobile application that analyzes, in real time, the data received by an external sonar device. Its main function is to detect the existence of obstacles in the path of the user and to emit information, through a voice interface, about the located distance, size and the potential motion and to advise as well how the user can avoid them.…”

Design of Audio-Augmented-Reality-Based O&M Orientation Training for Visually Impaired Children

“…The proposed solution is to use an external device that has an ultrasonic sensor and a GPS module to analyze the data received by the external sonar device in real time and decide whether it needs to emit an audible warning to the visually impaired person. The proposed algorithm can be equipped with a direction and navigation device, allowing the visually impaired person to walk safely and autonomously outdoors [ 13 ]. The disadvantage of the device is that the ultrasonic program cannot build a map and the ability to sense real-time changes in the surrounding area is poor.…”