>> By far the acceleration/deceleration of the car relative to the
>> earth's surface/road is what an accelerometer will measure.

> Totally wrong.  It will also measure 1g up, the apparent 
> accelleration due
> to gravity.  In other words a "stationary" accellerometer sitting on 
> my desk
> will measure about 9.8m/s**2 upwards.

He qualified his comment with the cavea "perpendicular to the earth's 
surface", which is liable to be the useful direction for a single axis 
accelerometer to be aligned when attempting to measure braking.

He was, if I understand him correctly, incorrect to the extent that 
the surface slopes relative to a line from the centre of the earth - 
ie local vertical. As I know you are aware, on  a hill a horizontal 
accelerometer will measure Sin(X) * g where X is the angle of the 
slope.

Putting together what several people have already said.

- Brake light signal will be useful.

- A fairly sure fire method to detect braking is to vector sum 
vertical and horizontal components relative to the car. With no 
braking these should always be ABOUT 1g in vector sum regardless of 
slope. If the sum exceeds one g it will be due to braking except that 
transient conditions (rate of change of acceleration/deceleration) 
will produce transient signals



        RM

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist