3d scanning with VL53L5CX

[ak_g]

Hello, 

I have started working with the VL53L5CX ToF sensor, for 3d scanning. I have successfully interfaced the sensor with MCU and am able to read the data from it. I have read the datasheet and um2884 user manual as well which explains the different available features of the sensor. A few things I am still confused about -

1. If I want to develop my own firmware for the sensor, I am not able to find the I2C protocol register addressing in the datasheet. I found a library from ST micro for interfacing. How can I make my own library if needed?

2. For a 4x4 or 8x8 resolution, what is the size of the smallest block in the grid at a particular distance x (in mm). I am asking about both the size of the whole grid and the size of the smallest block in the gird. This will make me understand what is the least count of the point cloud in my scanning.

3. What is the minimum thickness of a material that the sensor can detect. 

4. I am not able to understand the option called "Target order", and what is the strongest target order means, What can be a use case of it.

5. In the datasheet, the orientation of the sensor detection is mentioned. Is there a way to integrate the orientation in the code to make angle correction? Or we need to orient the sensor wrt pin A1 when scanning an object.

Looking forward for the response

akg

[John_Kvam · Accepted solution]

1. If I want to develop my own firmware for the sensor... ?

Alas I'm afraid that is not possible. If ST gave you that ability, there is no way they could guarantee the performance numbers in the data sheet. Besides the documentation on how to program it would be a nightmare they wouldn't want to deal with.

2. For a 4x4 or 8x8 resolution, what is the size of the smallest block in the grid at a particular distance x (in mm)... ?

A great question, but not an easy answer. A small spot of retro-reflective material (think safety vest) would be trivial to spot, but it would take a rather large piece of very black matte finish material. It has to do with the number of photons returned to the sensor. And that is a function of reflectivity and distance. Best I can offer, is to find some likely target and test that. 

3. What is the minimum thickness of a material that the sensor can detect?

Any material that reflects photons can be seen. An extremely thin bit of gold foil for instance would be thick enough. If the material were translucent - like most plastics, the material has to be thicker. But it's not the thickness as much as the reflectivity. 

4. I am not able to understand the option called "Target order", and what is the strongest target order means, What can be a use case of it? 

When the photons bounce off the target they reflect back to the sensor. The rate at which they return is the signal rate. The signal rate that works best is about 20Million Photons per Second - 20MPS. The rate is a function of target reflectance and distance. (The light dissipates as a square of the distance.) Most of the time the nearest target is the closest one. But not always. 

As to which is best depends on your situation. If you want to know the distance to a wall and don't care something small were to enter the Field of View, go the strongest. But if you want to detect that small target, go with the nearest. 

5. In the datasheet, the orientation of the sensor detection is mentioned. Is there a way to integrate the orientation in the code to make angle correction? Or we need to orient the sensor wrt pin A1 when scanning an object.

In order to get orientation of your target you are going to need to know where A1 is relative to the world. ST makes some handy sensors for detecting tilt - and gravity. 

- john

[ak_g]

wrt point #3, is there a value for minimum thickness, for different reflectance, some kind of graph?