Hi, I have a Matrice m100 with guidance attached in default configuration i.e. below the matrice base is battery and then the guidance sensor.
can someone help me with following queries/issues

1) Does guidance really use all 5 sensors to computer odometry?

Does it really compute stereo/mono odometry from all sensors and fuse them to obtain a best estimate. I have noticed it only using the bottom sensor. Other 4 horizontal

sensors seem to be used for avoidance only in which case also it seem to be using only the sonars and not stereo depth(I have started doubting if stereo pipline is even being

used as of now)

2) Can Matrice switch between P-GPS and P-OPTI modes when required (going under a bridge)?

So far in my experiments it swtiches to P-Atti when GPS deteriorates, I expected it to use OPTI in such a case because I flew matrice over textured surface.

It only switches to OPTI when flying close to ground with bad GPS and then directly goes to P-GPS when GPS improves but vice-a-versa never happens i.e. it never switches

to P-OPTI when GPS deteriorates. It only switches to P-Atti from P-GPS.

3) Which sensor governs the height when both GPS and sonar is available?

With guidance attached I expected it to maintain a better hovering height above ground but it seems to compete between GPS/Sonar/baro and ends up oscillating in height.

In all I believe everyone is excited to see a good system that Guidance is on Spec sheets and the website, but some things just dont work out. Maybe there is a beta firmware

that enables the stereo odomtery etc. I beleive so because we dont get access to all the disparity images to do our own odometry hence, there really must be something

onboard in development.

hey geetes, yes ours does automatically switch to opti when going under cover without gps..
I had your problems a while ago, but they released new firmware on the guidance shortyl afterwards and i did a recalibration , this was about 3 weeks ago, no problems since.
maybe just re-calibrate every camera axes .
its rock solid indoors. it seems to rely on the down sensor at first but eventually stabilises.

to test it indoors rather turn off the obstacle avoidance, it causes oscillations.

jasper Posted at 2015-9-12 11:55
hey geetes, yes ours does automatically switch to opti when going under cover without gps..
I had yo ...

Hey Jasper, thanks for sharing your experience . Guidance was on the latest firmware and I had recalibrated for the experiments I did just before starting this thread . I will do the calibration again. Regarding that, did you use a BIG screen to calibrate the guidance? I will retry

jasper Posted at 2015-9-12 11:55
hey geetes, yes ours does automatically switch to opti when going under cover without gps..
I had yo ...

Also, is your guidance setup in default configuration with following order going down
(up) Matrice Base -> Battery -> Guidance (down)

I remember reading somewhere to shift the battery above the base to reduce noise on guidance.

1. We indeed use all the 5 sensors to estimate states of the UAV.
2. Matrix 100 is suppose to switch to P-OPTI mode if GPS signal deteriorates, but for your case, I guess there is no texture on water surface or circumstance is too dark.
3. We control the height of UAV with respect to the HOME point but not the ground.

rdteam11.dji Posted at 2015-9-29 04:48
1. We indeed use all the 5 sensors to estimate states of the UAV.
2. Matrix 100 is suppose to switch ...

Hi, is there a way to tell if guidance is using only downward camera or all cameras?
I think if there is texture on sides but not enough texture on ground, then it should still be able to use OPTI. Is that right.