June 2006, Issue 191

Nontraditional Cursor Control
ATmega32-Based Motion Sensing

PROGRAM DESIGN

Figure 2 is a block diagram of our program. First, we wait for the computer to toggle the RTS line. When it does, we send it the “MZ” message. We then respond to five queries, disable query response, and start mouse functioning.

After the mouse has started functioning, the buttons are sampled on every cycle. If the mouse isn’t enabled, the program cycles through the button sampling until the mouse is enabled. After the buttons are sampled and the mouse is enabled, a check is done to see which acceleration axes have been sampled. If X has not been sampled, the sample multiplexer control variable is set to Y, the ATmega32 microcontroller sleeps until X is ready, and X is sampled when the ADC has finished processing.

The multiplexer is set to Y before X is sampled because the current sample being processed by the ADC is always taken off of the current multiplexer value. As such, if we want the ADC to sample Y, we have to set the sample multiplexer control variable to Y before the desired sampling period comes around.

If X has been sampled, we check to see if Y has been sampled. If it hasn’t been sampled, then the mux control variable is set to X, the microcontroller sleeps until Y is ready, and Y is sampled when the ADC has finished processing. If Y has been sampled, the ATmega32 microcontroller modifies the samples to the correct two’s complement numbers, scales them, and creates the serial mouse data packets. Then, the serial packets are sent, the mux is set to indicate that X needs to be sampled, and the program returns to the beginning of its sampling loop. The mouse is enabled and disabled by one of the sampled buttons. If the button value is read as pressed and its previous value was not pressed, the mouse enable/disable is toggled.

The tricky parts of this program were the timing and the conversion of the accelerometer output to the proper two’s complement number. We had some problems with the accelerometer readings. They became erratic on certain types of reads, specifically when we moved the mouse on the NW-SE plane. At first, we couldn’t figure out how to get steady readings. But then we found a web page addressing the problem that had been put up by students who had taken our microcontroller programming class. Those students had performed similar sampling operations. To get good readings, they put their microprocessor in Sleep mode until the result was ready.

We tried simply adding sleep times immediately after we began our ADC conversions. The entire setup then worked as we expected it to. Some post-change information gathering told us that to achieve the highest levels of accuracy with the ADC, you have to put the microprocessor in Sleep mode during conversions. We found that to be interesting.

We formatted the numbers by performing mathematical operations on them. Our math often caused unexpected overflows though. These overflows, coupled with the fact that we could never tell if we were casting the numbers properly, prompted us to devise an almost purely bit-manipulative method of converting the accelerometer outputs to numbers that we could send to the computer (see Listing 1).