“Noise Canceling” Human-Machine Interfaces

People are reliant on machines in nearly every facet of daily life. Some machines help wash clothes, others transport us between locations, and more yet, give us access to the vast resource of the internet. A specific kind of machine, appliances that function via a human-machine interface (HMI), are especially useful for niche tasks that the majority of us might take for granted. Disabled and non-disabled individuals alike are reliant on these HMI mechanisms to perform certain tasks, but discrepancies in the input, like background “noise”, can impede their utility. 

Human-machine interfaces (HMIs) are defined as mediums that allow people to interact with machines and systems. Some HMIs can be implemented with auditory input, others with tactile (touch-based) input, which means it receives stimulus from the environment and translates it to a device. However, many times the stimuli processed by the machines can be convoluted by unwanted background noises, rendering the output to be muddied. Recently, researchers have worked to create HMIs that would function without input from these background noises, making for a clear, more efficient process.

An artistic and literal rendition of Human-Machine Interfaces (HMI)

Engineering can take on many forms, and what the researchers engineered to overcome the difficulty of unclear inputs uses several composite materials that are sensitive to electricity, produced from specific stimuli. By really honing in on the creation of this structure, the researchers were able to form “frequency-selective” auditory sensors and more-sensitive haptic (touch-based) sensors that can eliminate the muddied input that had previously been a problem. By specifying what qualified as input, the researchers were able to produce an HMI voice-recognition system that was 95% accurate, as well as a touch-based system that could differentiate between surface textures. 

Auditory and Haptic Sensory Input Visualization  from the research

What is even more interesting is that these individual inputs, auditory and haptic, were then combined to control a robot. Typically, with excessive interference, such tasks would be very difficult. Controlling a robot through indirect means, like an HMI, can easily lead to distracting input, but with the introduction of this new technology, the background “noises” that had once posed a significant problem in signal interpretation have been quantitatively reduced, and the results are overall indicative of growth and maturation of the technology as a whole. 

The effects of this finding on “humanoid robots, wearable devices, and biometrics” are promising, and as HMI technology progresses, we can only hope to imagine what benefits such an efficient translation system can bring in terms of better lives for those whose struggles are often overlooked.