Research And Application Of Electronic Blind-guiding System

With the development of embedded,big data and artificial intelligence,various research and inventions related to ETA(Electronic Travel Aids)have been proposed,and there are two critical questions in the development of the application of ETA: The first question is how to collect comprehensive,detailed and accurate information from the surrounding,with using algorithms to convert raw data into information about obstacle avoidance and navigation which blind people need to know.On this issue,ultrasonic sensors,infrared sensors,machine vision and image identification,RFID,laser radar,and many other data accesses and algorithms are involved in the research of ETA;The second question is how to inform the collected information to the blind.It is necessary to ensure that not only the highly refined and accurate information is get by the blind,but also the blind gets the information accurately and efficiently.On this issue,ETA research mainly develops in two directions: auditory feedback and tactile feedback.Therefore,the main research contents of this paper are as follows:(1)This paper introduces the background and necessity of the research,and then finds out the key questions in ETA through the comparison and analysis of previous relevant studies,which will serve as the starting point and innovation point for the follow-up work.Then it introduces the basic principles needed for the design and construction of the system,which lays a foundation for the arrangement of sensors,design of obstacle extraction algorithm,ranging error correction,stereo auditory feedback and other work.(2)Based on infrared sensor and ultrasonic sensor,this paper designs and implements an electronic guide car for the blind,which managed to complete obstacle avoidance and guidance using low cost ranging sensors.Firstly,in the structure of the guide device,this paper adopts the design of combining an electrical car and a blind stick,and reasonably arranges 7infrared sensors and 2 ultrasonic sensors to detect the potential obstacles in all directions.Secondly,this paper implements range corrections based on the ranging principle and odometer,and obstacle avoidance and prompt algorithm combining with actual application scene.This paper also implements a function which can get wide range data from surrounding by using the infrared sensors scanning the surrounding when the user changes the direction of the guide car.Besides,his paper implements the data communication between electronic guide car and Android.Finally,an experiment including three walking tests under real environment is designed and implemented.The results show that the electronic guide car can comprehensively detect obstacles in all directions and heights,and provide effective obstacleavoidance guidance to the user.(3)In allusion to the problem that the people with impaired or no vision can't comprehend the surrounding spatial information detected by sensors,an auditory spatial perception blindguiding system based on laser scanner and Unity3 D is proposed.Building on the theory of Interaural Time Difference,Unity3 D is used to create virtual reality environments and simulate the surrounding where the data is collected by LIDAR at the frequency of 4000 Hz in the real life.Through the algorithm,the prompt center point of the obstacle is calculated.With the dynamic scene model at 50 FPS,the immersive stereo sound to represent the information regarding objects angular orientation,and varying pitch and loudness to represent horizontal distance,an effective auditory spatial perception can be created by the system for the user.An experiment with elaborate design is implemented in this paper,and the results demonstrate that the participants equipped with the system can effectively identify the orientation,distance and movement of multiple obstacles at the same time.