View Full Version : Vertical movement sensor

08-05-2018, 01:08 PM

I am wondering if anyone knows a sensor that only detects a vertical up and down movement?

I came across a sensor that could work for my application but it's very expensive. https://www.clrwtr.com/Products/B2N45H-Q20L60-2LI2-H1151

08-05-2018, 06:39 PM
No, that sensor does not actually measure up-down movement. It measures inclination about one or two axes, so basically gives you the same information as a plumb bob attached to one or two protractors.

You could get away with a single axis accelerometer if you can guarantee the sensor (and whatever it is attached to) will not be rotating about any axis (chassis rotating upward to drive up a hill). You could also try a very sensitive altimeter/pressure sensor; some of them have precision down to 1cm of air, but that makes them sensitive to interference from local airflow near the sensor.

Your most reliable method of easily measuring vertical/altitude change is with a 9DOF IMU/AHRS sensor (like the LSM9DS1) to feed a sensor fusion filter (like the Madgwick filter (http://www.x-io.co.uk/category/open-source/)) and feed the resulting orientation quaternion plus the sensor's linear acceleration and rotational velocity into an odometry system then ignore everything that spits out except the altitude.

08-05-2018, 07:06 PM
How far up/down do you need? A barometric pressure sensor could get you approximate up/down movement.
Of course, it would also tell you you went "up" just because a high pressure front started moving in, so it depends on environments.

Is the location fixed? If so, any distance sensor, such as ultrasonic, infrared, or time-of-flight sensor can get you what you want. This could work for non-fixed things, too, if you need "height above whatever is below me."

Other than that, I agree with tician, just get a 3DOF or 9DOF IMU and read out only the one of the axes you actually care about. (Or, if your sensor is rotated, use dot product to project the axis you care about.)