Supported LiDAR sensors

[LarteyLawson]

Hi Janusz,
Must the supported sensors have IMU??
I am using a Velodyne VLP 128 Alpha prime and a little curious if the sensors must have inbuilt IMU. The recorded file can be obtained in veloview and .laz frames can be exported based on the these frames. However, I without the IMU and other datasets can these frames be used aligned in with Mandeye system?? What do you suggest?? I can also share a sample file/folder for your own perusal. Thanks

[JanuszBedkowski]

Hi, we provide tools for other lidar edometry algorithms, I suggest to find most suitable solution by runnining benchmark. Somę of the algorithms do not require IMU.

[JanuszBedkowski]

ANother suggestion is to buy IMU, since it is plenty times cheaper than Velodyne VLP 128 Alpha prime.

[JanuszBedkowski]

Can You check https://github.com/michalpelka/kiss-lidarodometry

[LarteyLawson]

Thank you Janusz,
to be honest I still dont understand the benchmarking part. I have seen the videos and all but as to why and the bridge between the two , I am sorry I havent figured it out.
thanks

[JanuszBedkowski]

Did You check https://github.com/michalpelka/kiss-lidarodometry

[LarteyLawson]

yes. it crashes after I load the data and selecting ICP. During the data loading, it does not see the point data in the laz files. it reads only zeros for number of point in file.

"###
compressed : 1
file 'E:\GGScan_hpc\Alpha prime\Potsdammer\laz_output\lidar_700.laz' contains 0 points
number points with ts == 0: 0
counter_filtered_points: 0
total number points: 0
number points with ts == 0: 0
counter_filtered_points: 0
total number points: 0
0
total number points: 0
0
counter_filtered_points: 0
total number points: 0number points with ts == 0: 0
counter_filtered_points: 0
total number points: 0
total number points: 0
number points with ts == 0: 0
counter_filtered_points: 0number points with ts == 0: 0
counter_filtered_points: 0
total number points: 0
###"

[LarteyLawson]

Sample files attached through this link

https://drive.google.com/drive/folders/1ei-K1ckDUEj4b3iQMtgEgl0VrN9pCh12?usp=sharing

[JanuszBedkowski]

Hi, it is similar case to KITTI -> we are working on such data, I hope having some update that HDMapping-LIO will work without IMU, so it will be HDMapping-LO

I will back to this issue ASAP

[LarteyLawson]

Thanks Janusz,
After spending sometime on this issue, i wanted to brief you what i realized worked and maybe ask for all future velodyne users.
LidarView 5.2 works well and allows for the exporting laz frames. The internal Kitware Slam within LidarView also works decently but doesnt give one all the control like Mandeye HDmapping Step 2 does. It generates this ".poses" file from the trajectory.

Time,Rx(Roll),Ry(Pitch),Rz(Yaw),X,Y,Z
78.11018799999999374,0.00000000000000000,-0.00000000000000000,0.00000000000000000,0.00000000000000000,0.00000000000000000,0.00000000000000000
78.21021199999999851,0.00051131403503223,0.00019551791625188,-0.00092941518334166,-0.00023761726829178,0.00686586984228643,0.00136500463402859
78.31020499999999629,0.00012255423267036,0.00036504949912417,-0.00162907659191322,0.01876309698601771,-0.02248805840445888,-0.00161071750244157
78.41020000000000323,-0.00099727846107563,-0.00005711308838336,0.00110281425936839,-0.02066177696337781,0.02938742341926123,0.00817001606684813
78.51025599999999827,0.00049495337651571,0.00066814640607220,-0.00152857777679821,-0.02572312451833780,-0.00574956927867568,-0.00763336713628906
78.61030900000000088,-0.00101867996511067,-0.00009245674442095,-0.00237051277426115,0.06587384503910210,-0.03981925307702596,0.00567692621796033
78.71030700000000024,0.00091701031377228,0.00015092560493908,-0.00144284568911534,-0.00649680916437857,-0.00598740118049346,-0.00484734863401449
78.81032999999999333,0.00020075036739976,-0.00014029329310418,-0.00196704177508076,0.01739170262187472,-0.02083339424771067,-0.00014724612237575

Now a potential work around option, is Mandeye framework accepting the frames of Laz file and this type of trajectory.poses file .
for at least step 2. this is a bridge I believe is needed.

Bests