TL-4319
commented
15 hours ago I made some changes to my raw cloud to bag converter. Instead of writing the intensity values as uint32_t based on blickfeld definition, I wrote it as float. The new bag is as follow.
Now I am seeing no error regarding unrecognized field but I just not detecting the board at all
My suspicion is that the value of intensity is somehow now set correctly and the intensity filter with bound [0 15] (based on definition in line 82 of file AutoDetectLaser.h. Based on rviz, my max intensity range is [0 ~4000] which kinda doesn't make sense. I think my understanding of the intensity filter is wrong. Please let me know how I can scale my intensity to not have the entire pointcloud rejected and how to debug this problem.
Thank you
Hi, gain thank you for the great tool. I am beginning to test the tool with my custom sensor suites where I logs everything as raw data and then construct a bag later on. The LIDAR that I have is blickfeld cube which is a MEM scanning mirror. The scanning pattern is repetitive and is typically visualized as shown below
I was able to test my thermal and RGB images with LVT2 tools and is currently struggling with the LIDAR. Below is a visualization of the raw pointcloud and error
bag file of this dataset
The error said that it fails to find match for field 'intensity', does that mean my point cloud is missing the intensity field? I have confirmed that my files does have intensity. The point field is shown
I would like to get some guidance of how to get the tool to detect the board. I have tried using the robosense_m1 launch file since they are somewhat similar being repetitive scan sparse lidar and I don't have all information on how to convert blickfeld to velodyne type cloud.
Thank you in advance