I have a coworker who claims we can monitor when a vehicle is moving using an accelerometer and bet me $20 when I said we can not.
I understand that we can determine acceleration and deceleration but do not think we can detect when the vehicle is moving at a constant speed, especially when we have to build in some way to desensitize readings to accommodate vibration.
I win, right?
That’s somewhat correct. If you have a good enough accelerometer, you could integrate its output and calculate the current velocity.
This will not be feasible with a <$50 accelerometer though.
There are other ways to interpret that challenge. For example, we know that the car vibrates a lot more when moving than when stationary. Thus by characterizing the vibrations, you can have a pretty good idea of when the car is moving. Combine this with clever analysis of the accelerometer data, you can pretty reliability determine if the car is moving or stationary.
If you have a 2-axis accelerometer, you can compare the acceleration detected sideways from that in the forward/backward orientation. That could provide some clues. Furthermore, it is difficult to maintain a perfectly constant speed, given variations in road conditions and such.
Additionally, if you have a z-axis accelerometer, that could yield even more data. A stationary car has very little vertical movement/acceleration. But as the car moves over a surface, you start to see a lot more vibrations in this axis. This too can be used as indicator of movement.
If you want to be really creative, you could, for example, attach another accelerometer to the acceleration pedal so that you can know when the pedal has been depressed. This, in conjunction with the vibrations (which can be used to approximate the RPM), and the above said analysis, could theoretically offer a pretty reliable guess as to the state of the vehicle.
Daniel
Yes, it is possible:
http://en.wikipedia.org/wiki/Inertial_navigation
As monstrum said, it takes a good accelerometer, but you also need very accurate gyros (eg laser ring or fiber-optic gyros). Even good Inertial Navigation units can have a position drift of about 1 mile per hour.
MichaelN:
Yes, it is possible:http://en.wikipedia.org/wiki/Inertial_navigation
As monstrum said, it takes a good accelerometer, but you also need very accurate gyros (eg laser ring or fiber-optic gyros). Even good Inertial Navigation units can have a position drift of about 1 mile per hour.
I think that goes outside of the bet, using other sensors. I mean, if you are going to add other sensors, just add a GPS. 
Is it possible with the accelerometer alone? My answer would be no as well, but my theoretical knowledge is not strong here.
do I lose … I’m in a car moving to the east (no turns) and a constant 20mph … other than gravity and vibrations, what measurable acceleration is there?
"Let me tell you a story. When I was a young man about to go out into the world, my father says to me a very valuable thing. ‘Son,’ the old guy says, ‘I’m sorry that I am not able to bankroll you a very large start. But not having any potatoes to give you, I am going to give you some very valuable advice. One of these days in your travels, you are going to come across a guy with a nice brand new deck of cards, and this guy is going to offer to bet you that he can make the Jack of Spades jump out of the deck and squirt cider in your ear. But, son, do not take this bet, for if you do, as sure as you are standing there, you are going to end up with an ear full of cider.’ "
-Sky Masterson from Guys and Dolls.
Is this a sucker bet or a real bet?
I can think of a dozen ways to use an accelerometer to figure out if a car is moving without even looking at its output or connecting it to power. Just drop it out the window. If you look down and see it, the car is not moving.
dogfuel:
do I lose … I’m in a car moving to the east (no turns) and a constant 20mph … other than gravity and vibrations, what measurable acceleration is there?
IMO:
It’s a time issue: for the first x amount of time, you can estimate speed with just a simple accelerometer. After a while, it’s quite uncertain, due to drift and double-integration. Car NAVs (good ones), use a magnetic compass for heading, short term, and the vehicle’s speed sensor (drive shaft speed pulses or equiv) plus GPS speed/heading as inputs to a Kalman filter to get a good estimate.
Gee-wiz aircraft, etc., use fiberoptic gyros (based on speed deltas of photons in coils of fiber).
If the question is: Is Vehicle Moving? I’d use a proper kind of vibration sensor. e.g.,1/2 in. long, 1/8 in diameter cylinder containing a gold-plated ball (ball-in-tube). Cylinder’s walls are ribbed with conductive material. Lots of vendors make these.
Inertial navigation units are used where GPS isn't an option (eg underwater) or where navigation is critical and can't just depend on GPS (eg military applications). In certain military applications, it is possible to jam GPS signals to deny the enemy access.TheDirty:
I mean, if you are going to add other sensors, just add a GPS.
Gyros can also be used to improve GPS navigation systems on cars - when going through tunnels or other environments where the GPS signal is blocked, combining gyro information with the odometer can maintain fairly accurate positioning for moderate distances.
This is your answer. You win.Valen:
With just a single-axis measuring along the forward/backward directionr not. If the accelerometer is switched on while you are allready moving then there is no way to determine wether or not (or when) you are standing still. Since any deceleration from then on will be equivealent to accelerating from standstil in the opposite direction. Basic Newton’s laws of motion.
You guys need to step back from being technical nerds here and stop trying to find technical loopholes and added components to get the results.
You guys need to step back from being technical nerds here and stop trying to find technical loopholes and added components to get the results.
But then somebody whould take our pocket protectors away! We would be disgraced!
But the real questions was this a real bet or a sucker bet. And if it was a real bet what where the paramters. If the car was in steady state on an ultra smooth surface, than an accelerometer used in a conventional way would help. But tie a parachute to it and stick it on the roof of the car and that would give you a good indication if the car was moving.
The bet is too open-ended to say if you win or lose. In theory, and if you are allowed to switch on your accelerometer-only-device during zero velocity, then yes, you can determine if you are moving or not. Practically, yes, but only for a certain time (which depends on how much money you have).