Microcontrollers and human-machine interfaces

When I wrote about gesture interfaces two years ago, the Apple iPhone, Nintendo Wii, and the Microsoft Surface were brand new version 1.0 products. It was not a certainty that any of these products would be commercially successful a few years after their roll out. Two years later, all of these products are still available commercially, and in fact, they have moved well beyond version 1.0 into version 2.0 or even 3.0 versions. The iPhone operating system is version 3.1 and the product has incorporated many lessons learned into the system and software. The Wii MotionPlus remote supports higher levels of precision and accuracy than the original remote to better capture the user’s body and arm twists. The Surface continues to deploy with its latest announced location at Harrah’s in Atlantic City.

The ability to use software to incorporate lessons learned about the advanced human-machine interface that each of these types of products employs is essential for them to adjust to world-wide user needs and preferences. Better sensing technology coupled with increasingly sophisticated interface algorithms enable these devices to accurately and consistently interpret a user’s intent across a wide range of operating conditions.

Many semiconductor companies have offered touch interface technologies for years, but an interesting and significant trend has been expanding over the past two years – especially since the launch of these three products. More than a dozen semiconductor companies have been repackaging and expanding their touch technology offerings just over the last year or so. These offerings no longer consist of just hardware capabilities, but rather, they are wholly integrated solutions that include not only the sensor circuitry integrated with a microcontroller, but they include bundled software that provide increasing amounts of intelligence on how to read the sensors, filter out noise, identify unintended inputs (such as pressing your face against the touch surface when talking on a phone), and detect, manage, and interpret multiple, simultaneous touches to support complex gesture recognition including pinching.

The integrated microcontroller and bundled software is playing an ever increasingly important role in these advanced sensor offerings. Mind you, the instruction set architecture of the microcontroller is not the differentiating feature anymore; rather, it is how the microcontroller is integrated with the sensing circuitry, and how the bundled software exposes and abstracts the raw input data from the sensing array. The semiconductor provider is integrating more expertise in their software so that designers can use the sensing package more quickly and effectively than ever before. Companies that continue to rapidly incorporate lessons learned from their customers into their bundled software will probably be the winners in this emerging capability for embedded applications.

I had an interesting conversation, available as a webcast, with Trevor Davis from Cypress Semiconductor and Fanie Duvenhage from Microchip. We talked about the microcontrollers, bundled software, and supporting disciplines needed to build advanced human-machine interfaces. Please take a moment to check it out. We are answering questions via email. I would love to get your feedback for targeting a narrow scope and deep dive into specific aspects of advanced human-machine interfaces that interest you most in future webcasts.