Sharing issues and requesting technical advice regarding Garmin LIDAR-lite v4

Sensors Distance
1

Hello. I am a customer using your Garmin LIDAR-Lite v4 product. I have encountered the following issues during use:

a. When measuring the distance to a flat object 80cm away from the LiDAR, there is no distance error.
b. Keeping the object in place, I raised the height of the LiDAR by 10cm increments, setting it to point downwards to detect the object, and measured the hypotenuse distance from 0 to 50cm. The error gradually increased, and I observed an error of approximately 7cm when the height was 50cm.
c. I confirmed that there was no error when the LiDAR and the object were parallel. However, when I oriented the LiDAR to be perpendicular to the floor (pointing straight down) and kept the object parallel, an error of 10cm occurred. (In both cases, the distance was 50cm and the object’s material was the same.) Therefore, the geometrical issue of the LiDAR beam traveling diagonally relative to the table is ruled out as the cause of the problem.
d. The biggest issue here is that there was a 10cm error when the LiDAR was set to be vertical to the floor, but there was no error when it was vertical to the ceiling. This raises suspicion of a posture problem caused by gravity.
e. However, the most confusing point in this issue is that when I applied 3.3v to pin 1 (Vin) of the LiDAR, the error almost disappeared. (As everyone knows, the specifications recommend applying 5v.)

So, ultimately, I have three questions:

**1. Is 3.3V power supply the correct one and the specifications are wrong? (In fact, there are pages in the specifications that show both 5V and 3.3V.)

  1. What is the technical reason why the distance value changes simply by changing the power supply?

  2. Is the cause of this problem truly due to the difference in power supply? Or is it an illusion? If it is an illusion, what is the real cause of this problem?**

Thank you for your prompt response.

Thank you.

2

When the Garmin LIDAR-Lite first came out, I read the company’s (not Garmin) description of how their invention worked . The details may have changed somewhat with later versions, but based on the originally described method, I’m not surprised by your observations.

The original device worked by sending a shaped pulse and determining the round trip time, which requires detecting the center of the reflected pulse. That was done by pattern matching the incoming data using cross-correlation of the transmitted versus the received pulse envelopes, which is very far from trivial at light speed.

Reflection of the pulse by a surface that is not perpendicular to the incoming beam will necessarily distort and weaken the pulse (due to pulse spreading and different distances between reflecting spot extremities), which in turn will introduce errors into the recognition process and subsequent distance measurement.

I confirmed that there was no error when the LiDAR and the object were parallel.

All measurements have errors, but measurements from surfaces perpendicular to the beam should be more trustworthy than ones from slanted reflecting surfaces. Note that the expected errors depend on the distance to the target.

For reference, here are the technical specifications from the LIDAR Lite V4 operating manual.

Capture
Capture807×212 28.1 KB

This raises suspicion of a posture problem caused by gravity.

Extremely doubtful. More likely, the reflecting surfaces do not have the same reflectivity, and it is almost certainly lower than 90%.

(I’m not a Sparkfun employee, just an interested customer and user of a LIDAR-Lite V1.)