2024-08-27 10:39 AM
Hello,
I am using a VL53L8CX sensor in 8x8 mode. There is a thing that is unclear to me from the documentation, so I need a clarification. A value in the zone Z of the matrix I get from the ToF sensor is:
a) Distance between the object in the zone Z and the optical center of the sensor
b) Vertical distance between the object in the zone Z and the sensor plane
Which one of these is true?
Kind regards,
Milan
Solved! Go to Solution.
2024-08-27 11:37 AM - edited 2024-08-28 07:08 AM
It's b Vertical distance
Perhaps I can explain by telling you how the sensor works.
Somewhat simplistically the Vertical Cavity Surface Emitting Laser (VCSEL) send out a flash of light. Due do an optical diffuser the light goes out in a Pyramid of 45x45 degrees.
The light hits something and reflects back. The light goes through a focusing lens and is detected by an array of Single Photon Avalanche Diodes (SPADs).
The photon strikes are binned by which clock cycle they arrived in. Each bin is worth 3.75cm of distance.
By looking at how many photon strikes arrived per clock cycle we can determine how far away the object.
In reality, there are about a million flashes per range. And we like somewhere between 0.5 and 20Million photon strikes per second.
All this to say that you are seeing the distance from the optical center and the middle of the zone - almost.
But that's not what anyone wants so we go through a Radial to Perpendicular conversion.
This allows you to point the sensor at a wall exactly perpendicular and get 64 distances that are nearly identical.
- john
2024-08-27 11:37 AM - edited 2024-08-28 07:08 AM
It's b Vertical distance
Perhaps I can explain by telling you how the sensor works.
Somewhat simplistically the Vertical Cavity Surface Emitting Laser (VCSEL) send out a flash of light. Due do an optical diffuser the light goes out in a Pyramid of 45x45 degrees.
The light hits something and reflects back. The light goes through a focusing lens and is detected by an array of Single Photon Avalanche Diodes (SPADs).
The photon strikes are binned by which clock cycle they arrived in. Each bin is worth 3.75cm of distance.
By looking at how many photon strikes arrived per clock cycle we can determine how far away the object.
In reality, there are about a million flashes per range. And we like somewhere between 0.5 and 20Million photon strikes per second.
All this to say that you are seeing the distance from the optical center and the middle of the zone - almost.
But that's not what anyone wants so we go through a Radial to Perpendicular conversion.
This allows you to point the sensor at a wall exactly perpendicular and get 64 distances that are nearly identical.
- john
2024-08-28 02:35 AM
Dear John,
Thank you for your quick response. So, if I understood it correctly, although the sensor measures radial distances, the FW inside the sensor performs the conversion to perpendicular, effectively giving us option b? This would be in accordance with my observations where I pointed the sensor straight at the wall, and this is what I got:
[719, 720, 723, 722, 721, 717, 718, 715]
[715, 717, 725, 728, 725, 722, 721, 721]
[695, 722, 725, 729, 727, 727, 729, 725]
[721, 725, 730, 729, 729, 735, 735, 730]
[720, 730, 728, 729, 735, 735, 735, 734]
[722, 732, 733, 735, 735, 733, 731, 730]
[725, 724, 731, 735, 732, 738, 735, 732]
[734, 731, 727, 736, 734, 732, 737, 738]
I expected the distances to be increasing from the center towards the edge, but they are more or less the same over the whole field of view.
Kind regards,
Milan