VL53L8A

[xukong]

Can I ask you about the VL53L8 series.
Can the VL53L8 series use multiple at the same level at the same time, the number is greater than 4, will the middle affect each other, I may need to go through several 8*8 to form a larger depth map? Is the distance measurement in the TOF depth map obtained by this sensor in the middle of each small matrix?

[John E KVAM · Accepted solution]

The do interfere a little bit. If you place them directly at each other, or in the same direction looking at a particularly reflective object you will find some interference. 

but if they are close and pointed at different directions you should be just fine. A little light at the edges of one sensor might slightly distort the one edge of the one next to it, but the effect will be hardly noticeable. 

if you really are worried about it, start them at slightly different times. 

Start the even ones, then the odd ones, then try to keep up with all that I/O. Extra points if you can put them on separate I2C busses. But kicking them off at slightly different times, they cannot be in sync. And that will work to your advantage. 

I'm not quite sure I understand your other question. The sensors do a Radial-to-Perpendicular calculation. So if you put then perpendicular to a flat wall, all the zones will return the same distance. (Even though the light travels farther at the edges.)

So you can develop a point cloud using this data. There is some code on this community page:
VL53L5CX Multi-Zone Sensor get x,y,z of points rel... - STMicroelectronics Community

(The L5 and L8 have the same chip geometry, so don't worry that it says "L5". It still works.

[xukong]

Hello, we currently, due to the need to measure smaller objects at a distance (4M) (this object may only be 10cm^2, or even smaller, and may not be horizontal), consider whether it is possible to form a higher resolution depth map through multiple L8 sensors. But we don't know for sure, assuming that L8 is not resolution enough and the object is small enough, can we know the distance of the object?

[John E KVAM]

It's possible for us to build a part that has higher resolution, but it comes at a cost. And that cost is distance.

The sensor needs a number of photons returned in order to detect it. If the part is dull, or far away, we cannot see it. 

If we increase the resolution, we reduce the number of photon detectors per area and then we cannot see as far. 

Unfortunately turning up the laser power is not possible as we would cause eye damage. 

I'm thinking using Direct Time-of-Flight is not an option if you need that kind of distance and resolution. 

At least not yet. You are not the only one asking. 

I'm thinking stereo cameras would work. But they are a bit expensive. 

I also think the ST VD55H1 Indirect ToF would work. I know there is a databrief on ST.com. But I don't think the part is out yet. 

[xukong]

Hello, thank you for your reply.
I might want to try something new with the VD55H1, provided it's not so expensive and not so out of stock.
Stereo cameras may be a bit expensive and should not be considered for a while.