Low cost height sensor Lidar or ToF for detecting box heights on a conveyor

[azmathmoosa]

Hi,
I need some help in looking for a Lidar/ToF or any other height sensor for a barcode reading application. We have a machine vision camera with liquid lens whose focus can be adjusted to predefined focus planes. Now, if the height of the box is known, we can adjust the camera to the right focal plane and take a sharp snap. The conveyor is moving fast about 0.5 m/s and boxes can vary from LxWxH as 1mx1mx0.7m to .5mx0.5mx0.05m. The conveyor width is 0.8m and the box can be anywhere on the conveyor - not just center. Hence a single-point laser sensor won't work.

We need a sensor that can look at a slice of the conveyor fully covering the horizontal and output the value closest to the sensor. That way the box can be anywhere on the conveyor and we will focus correctly at the box height. The sensor should be cost effective.

 

[danadakk]

Maybe this - https://www.bananarobotics.com/shop/HC-SR04-Ultrasonic-Distance-Sensor

has 60 mS measurement cycle. You will have to do calculations to see if box remains in
measurement window long enough.

https://cdn.sparkfun.com/datasheets/Sensors/Proximity/HCSR04.pdf

Calculator - https://www.triangle-calculator.com/?what=iso&a=c=.8+C=15&submit=Solve


Regards, Dana.

 

[betwixt]

I doubt LIDAR would be cost effective.

Without seeing it my first approach would be simple light beams. Detecting height that way is easy, you just direct a light to sensors across the width of the conveyor and see which ones are broken and which are not. It gets a little more complicated if the boxes are different widths but the same height or can be at different positions across the conveyor.

It comes down to how accurately you need to focus the barcode reader. The lines on bar codes are arranged so the exact width of the code isn't too important so if you are using a standard scanning reader you don't need perfect distancing. I would try light beams projected inward from just outside the conveyor edge and up to sensors at a suitable angle that any box would shadow at least one beam. That way if you read all the sensor states you could use a simple look-up table to find the size and position of the box.

Ultrasonics is an option but you would probably face them picking each other up unless you sequenced them so only one transmitter was active at a time.

Brian.

 

[BradtheRad]

Construct a stack of photosensors as tall as your tallest box. Space the sensors 1 or 2 or 3 cm apart (whichever yields the precision you require). Place it on one side of the conveyor.

On the other side place a light box equally tall. Mount a stack of vanes (similar to a Venetian blind) in front, so that each photosensor sees no light except the beam emerging from the vanes which are at its own height.

 

[betwixt]

Brad, the problem is the boxes may be different widths and not always in the same place across the conveyor. Your idea works if only the height is variable. Adapting it to use slanted beams and shooting them across the conveyor from both sides allows the height and displacement to be measured. Its the same principle but applied in 3D space instead of 2D.

Brian.

 

[BradtheRad]

Brad, the problem is the boxes may be different widths and not always in the same place across the conveyor. Your idea works if only the height is variable.

Ah, ha... Right, I had the impression he wishes to measure height only.