The multi-touch capabilities afforded through use of capacitive touch screens have, over the last few years, proved incredibly popular in consumer electronics and computing designs. There are now numerous opportunities opening up for it in non-consumer applications too. Human machine interfaces (HMIs) supporting multi-touch interaction are able to detect all sorts of intricate user gestures, where otherwise the specifying of a more cumbersome form of HMI (such as a keypad) would be necessary - spoiling the user experience in the process. With multi-touch able to make the way people engage with technology less frustrating and more intuitive, there is a great deal of demand to broaden the scope of where capacitive touch technology can be utilised.
If capacitive touch technology is going to find its way into a broader spectrum of HMI designs, then engineering teams need to ensure it can be implemented in the most effective way possible. Figure 1 describes a normal multi-touch HM system - consisting of a high performance microcontroller (usually 32-bit), a large Flash memory for storing image data and a large capacity frame buffer through which this data is transferred to the display. Wide parallel buses are also needed to take care of communication between these ICs. The board real estate needed for such a system and the bill of materials costs associated with it have so far limited the adoption of multi-touch HMIs.
Figure 2 conversely describes a simpler, more streamlined multi-touch HMI set up based on FTDI Chip’s Embedded Video Engine (EVE) technology, which can determine up to 5 individual touch points simultaneously. Using the FT801 EVE device it has the ability to display bitmap images which the user can then easily resize through a two finger stretch/shrink gesture. The user can also change the orientation of such images via a multi-finger rotate gesture. There is plenty of provision for the determining of various other gestures - such as two finger taps, page flicks, cut-and-paste, swipes, and two finger long presses.
What differentiates this set up from conventional ones is that its audio and visual outputs all take the form of objects. This approach, which is the basis of EVE, allows the rendering of image content, for example, on a line-by-line basis (at 1/16th pixel resolution), instead of pixel-by-pixel (which is far more data intensive). As a result, it eliminates the need for a frame buffer or a large memory to storing images, and only needs low cost 8-bit or 16-bit microcontroller. The component count for this set up is far less than what is needed for conventional ones. It takes up less board real estate, shortens development times and requires a smaller engineering resource to complete. It is therefore much better suited to cost sensitive, space constrained implementations than conventional multi-touch HMIs solutions.
For more information visit: www.ftdichip.com/EVE.htm