Speed measuring device for low speeds (No GPS)

Thread Starter

jordiggie

Joined May 5, 2022
24
Hello everyone, I have started a new project where I am trying to determine the speed while mounted to a moving device measuring the ground. I have looked at a laser surface velocimeter but am unable to find one for small scale applications.

I would like to measure speeds between 0 and 10 km/h using the surface of the ground.

I have looked at using opto's interruptor and wheel but this can wear and tear over time so I am looking an other valid options before beginning.

Any input is appericiated!
 

Thread Starter

jordiggie

Joined May 5, 2022
24
Hi jord'.
Is this vehicle using wheels or tracks to move over the ground.?
E
No not necessarily. I have thought about adding a extension with a wheel that moves over the ground and a simple encoder to measure the rotations and thus calculate the speed.

I'm just checking if there are any other methods of measuring the speed/velocity through other methods.
 

MrSalts

Joined Apr 2, 2020
2,350
You can measure the rotation rate of one wheel (using an encoder directly on the wheel's axel or adding a measurement wheel against one of the vehicle's wheels - because that would be much easier than trying to maintain a fifth wheel in contact with the ground.
 

DickCappels

Joined Aug 21, 2008
9,297
Unless you have control of the surface - evenness, consistent markings, etc. It may be very difficult to come up with anything that beats a tachometer on a wheel with good traction.
 

Thread Starter

jordiggie

Joined May 5, 2022
24
Unless you have control of the surface - evenness, consistent markings, etc. It may be very difficult to come up with anything that beats a tachometer on a wheel with good traction.
Yeah I was thinking about adding springs to reduce being affected by uneven surface, but there has to be a better option then doing it like this. I know a tachometer is definitely the most easy on a wheel and the laser technology is way too expensive at the moment. I looked at GPS but since it would be indoors in a greenhouse it wouldn't be accurate either.
 

Thread Starter

jordiggie

Joined May 5, 2022
24
Here are some techniques that come to mind:

  1. ultrasonics
  2. radar
  3. lidar
  4. doppler
  5. imaging
I tried looking at doppler, but prices are out of my range in that area.
Not sure how I would apply imaging, but could be an option.

Im leaning towards eithher lidar or something in that area since I think it could work well, it's just out of my expertise. Would love to learn more about it and it's appliances so any sources would be very much appericated.
 

Thread Starter

jordiggie

Joined May 5, 2022
24
You can measure the rotation rate of one wheel (using an encoder directly on the wheel's axel or adding a measurement wheel against one of the vehicle's wheels - because that would be much easier than trying to maintain a fifth wheel in contact with the ground.
Yeah I have thought about it and it's kind of my back up option, just exploring other possible options before determining what I want to use.
It's definitely easy to implement and I know it will give accurate results.
 

MrChips

Joined Oct 2, 2009
27,112
I tried looking at doppler, but prices are out of my range in that area.
Not sure how I would apply imaging, but could be an option.

Im leaning towards eithher lidar or something in that area since I think it could work well, it's just out of my expertise. Would love to learn more about it and it's appliances so any sources would be very much appericated.
I am assuming that this is a DIY project. Hence you would not be looking at commercial solutions.

Doppler shift means looking at the scattering of a laser beam and measure the frequency shift. We did this in one of our Engineering Physics laboratory experiment. You can also do this with ultrasound.

For imaging I am thinking take a picture of the ground every 10ms. Use pattern recognition to measure displacement. This is how optical mouse works. In fact, that would be something to try. Take an optical mouse and see what you can make of it.

10km/h is about 2.8m/s or 28mm/10ms
 

Thread Starter

jordiggie

Joined May 5, 2022
24
Yeah I have thought about it and it's kind of my back up option, just exploring other possible options before determining what I want to use.
It's definitely easy to implement and I know it will give accurate results.
I like the idea of the doppler shift. Are there a lot of external tools that you need for this besides the laser beam? I'll go look in to this option.
 

Thread Starter

jordiggie

Joined May 5, 2022
24
Does that mean that you will always have a “target” (such as a wall) against which to bounce the signal?
Sadly since it's in a greenhouse the walls are made of glass which are lightly coated so the reflection will not always be perfect. I've thought about a optical laser measuring the distance and then when moving calculating the speed, but the reflection of the laser is not always the same leading to irregularities in the calculations
 

Thread Starter

jordiggie

Joined May 5, 2022
24
I am assuming that this is a DIY project. Hence you would not be looking at commercial solutions.

Doppler shift means looking at the scattering of a laser beam and measure the frequency shift. We did this in one of our Engineering Physics laboratory experiment. You can also do this with ultrasound.

For imaging I am thinking take a picture of the ground every 10ms. Use pattern recognition to measure displacement. This is how optical mouse works. In fact, that would be something to try. Take an optical mouse and see what you can make of it.

10km/h is about 2.8m/s or 28mm/10ms
What components did you use to measure the doppler shift?
And not sure if you can answer this question either but; When measuring the doppler effect, does it also work when I am measuring from a moving object to for example the surface? The received frequency will still be different then the one sent out right?
 
Last edited:

WBahn

Joined Mar 31, 2012
27,389
Hello everyone, I have started a new project where I am trying to determine the speed while mounted to a moving device measuring the ground. I have looked at a laser surface velocimeter but am unable to find one for small scale applications.

I would like to measure speeds between 0 and 10 km/h using the surface of the ground.

I have looked at using opto's interruptor and wheel but this can wear and tear over time so I am looking an other valid options before beginning.

Any input is appericiated!
There's some devils in the details here. How accurate do you need this to be? Do you just need a decent idea of the current speed, or are you going to be trying to use this in order to track the vehicle's position over time? What considerations are there when the vehicle is turning or maneuvering such that one part of the vehicle is traveling at potentially a considerably different speed than another?

If you are going to use the ground, what does the ground look like? How uniform and consistent is it? How smooth? How reflective?

How far off the ground would your sensor be located? How much might that distance change, either as the vehicle moves (rocks, bounces, whatever) or as the ground varies (small up/down features or objects/holes)?
 

Thread Starter

jordiggie

Joined May 5, 2022
24
The moving object is a distribution system for fertilizer in my greenhouse. I want to input the amout that has to be distributed per square meter and have everything else automated. So for instance if it is hanging still it doesn't distribute and when it is tugged on or pushed it measures it's movement and starts spreading out the amount put in by me.
 

Beau Schwabe

Joined Nov 7, 2019
129
imaging? .... Years ago (20 maybe) we had a bus driver that would fly through our neighborhood. I was able to video tape him driving between two markers that I knew how far apart they were. By incrementing the video frame by frame and knowing the frame rate, I was able to determine how fast the bus was going... almost 40 in a 25. I filled out a full detailed report and sent it to the school district. ... That took care of the bus driver. Now days this would be somewhat trivial with a raspberry PI
 
Top