Visually Impaired Pulse Echo Ranging (VIPER)

At least one blind person has developed a high level of proficiency in echolocation using tongue clicks. However, the clicking tongue is a single emitter. The VIPER device provides multiple emitters oriented in different directions to allow a mosaic image to be formed from multiple sectors of echo information, enhancing overall spatial perception.

Self-consciousness of making clicking sounds with one's tongue can discourage persons from making tongue clicks. The electronically generated acoustic pulses of the VIPER device can avoid such self-consciousness by attributing the sound to the device, not the user, similar to way those who are uncomfortable singing karaoke are generally comfortable listening to music from other sources.

http://www.bbc.com/news/magazine-19524962

Functional magnetic resonance imaging (fMRI) provides an amazing way to study the portions of the brain involved in various human activities in real time. Scientists have used fMRI to study human echolocation in early and late blind echolocation experts.

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0020162

Parts 1 and 2 of a video of the VIPER prototype have been uploaded to Vimeo. See details section above for links.

My Tek 2212 DSO isn't the latest technology, as can be seen from the sparse dots of the waveform, but, with a microphone connected to it, it does reveal that the pulses rapidly rise and exponentially decay, falling to half their peak level within 3mS. Sound travels at approximately 340m/s, but, since it has to go out and come back before the echo is received, its speed is effectively halved to approximately 170m/s for echolocation in terms of distance to the object being perceived. That means the approximately 3mS pulse effectively occupies about half a meter of echolocation range.

The pulse waveform oscillates at around 2.3kHz, perhaps as a result of a resonant frequency of the piezoelectric transducer. I have ordered more transducers and plan to employ them in different ways and possibly modify them to alter any resonances they may have.

Posted high-resolution photos of the project showing the VIPER prototype inside and out.

The uploaded files have the file extensions .net, .pin, .gpi, and .pro.

The NC drill files, with the file extensions .drd, .dri, and .whl, are Excellon NC drill files.

Note: The Gerber files, with file extensions .cmp, .sol, .plc, .pls, .stc, and .sts, are Gerber RS274X files.

Note: The piezoelectric transducer pads, power switch pads, and battery pads are laid out as if those components are to be mounted on the printed circuit board (PCB), but those components can (and at least in the case of the transducers should) be separated from the PCB by wires.

Note: The schematic diagram shows TN0604N3 MOSFETs as they were available in the parts library and have the same pinout as the 2N7000 specified in the bill of materials. The schematic diagram also shows piezoelectric transducers that were available in the parts library, but any of a wide range of piezoelectric transducers can be used. The capacitor and resistor values are not critical, and a wide range of values can be substituted.

The carefully tailored acoustic pulses emitted by the VIPER provide an abrupt wavefront that reduces ambiguity in spatial perception even if simply providing a higher signal-to-noise ratio of the returning wavefront. However, those pulses are also interesting in that there is a phenomenon known as the precedence effect, which describes a psychoacoustic masking of subsequent wavefronts after a first wavefront is perceived. The duration over which the masking persists is dependent upon the nature of the sound being heard, with a shorter duration of masking for acoustic pulses and longer duration for more complex sounds. Accordingly, the carefully tailored acoustic pulses of the VIPER provide advantage not only in perceiving the range of the closest acoustically reflective object, but also in resolving additional objects at greater distances.

Recent research suggests that a person's engagement in an echolocation task, as opposed to mere listening, can reduce psychoacoustic echo suppression. As the experiment distinguished between listening to exogenous sounds and engaging in echolocation using self-vocalized sounds, it appears some question may remain as to how echo suppression may be affected when engaging in echolocation using exogenous sounds. While control experiments were performed to attempt to distinguish effects of self-vocalized sounds vs. exogenous sounds, it is questionable whether the duration of such experiments were commensurate with the on-going duration of interaction of a user with the VIPER.

One question I have is whether a user's extended interaction with the VIPER can inhibit echo-suppression during the echolocation task even with exogenous sounds. One phenomenon that could conceivably support such improved performance over time is neuroplasticity.

Even if the precedence effect were not appreciably reduced, that is not problematic for the VIPER, as precedence effect still allows perception of the echo from the closest acoustically reflective object, which being nearest to the person using the VIPER, is likely a more important object of which to be aware than objects farther from the person.

http://www.psychologicalscience.org/index.php/publications/observer/2015/december-15/using-sound-to-get-around.html

http://rspb.royalsocietypublishing.org/content/280/1769/20131428