Lidar fault - due to unstable distance value


My Lidar lite v3 is connected to Pixhawk through I2C.
The binary is built by “make px4fmu-v2_lpe” with recently pulled git source code.
Lidar is successfully connected with simple change of “rcS” (ll40ls start pwm -> ll40ls start i2c) and able to see DISTANCE_SENSOR.current_distance value through QGC.

The problem is lidar distance value has too much noise.
Even not armed staystill position, it’s randomly fluctuate from 9 to 27 when distance between lidar and floor is 17.
As it’s unstable the PX4 log shows " lidar fault, beta XX.X" and “lidar OK” frequently.
INFO [local_position_estimator] [lpe] lidar fault, beta 17.68
INFO [local_position_estimator] [lpe] lidar OK
INFO [local_position_estimator] [lpe] lidar fault, beta 9.09
INFO [local_position_estimator] [lpe] lidar OK
INFO [local_position_estimator] [lpe] lidar fault, beta 10.99
INFO [local_position_estimator] [lpe] lidar OK
INFO [local_position_estimator] [lpe] lidar fault, beta 9.27
INFO [local_position_estimator] [lpe] lidar OK
Does anyone had same issue? or any idea how to fix this?
Is this only v3 problem? or I miss something.

FYI, lidar is powered by BCE.



I’m having a similar problem when using a TeraRanger. The distance measures look fine when viewing through QGC, but that same fault keeps appearing in the log.

Did you ever come up with a solution to this?

Thanks in Advance!



Well, I just lower my expectation for the sensor’s resolution. It seems the sensor’s output is
The parameter “LPE_LDR_Z” is for standard devication of lidar sensor which stands for it’s noise variation and this value determine lidar fault state based on sensor’s real input variation.
You may increase the value to suppress ths fault logs I’ve mentioned.
In my case, 0.08 works.
(remember, this doesn’t help to increase the lidar’s stability, but only changes fault decision level)

Hope this may help you.