Point Cloud Compensation (motion distortion correction)
I'm using ROS Noetic on Ubuntu 20.
I'm working with a VLP-16 LiDAR mounted on a vehicle that scans while the car is moving. From what I read, from papers and datasheet, that due to the linear and angular velocity when scanning, part of the reading is distorted and needs to be transformed to the current frame. (some points seem further than they actually are).
So before working with the point cloud, I need to correct it due to the distortion caused by the vehicle movement.
I can't find a package/function anywhere that does so, I'd prefer it to be in python but anything really would be beneficial.
In summary: I need a package/function that does point coud correction/compensation due to sensor movement while scanning.
Can you be a bit more specific about "motion distortion"? Do you want to transform the pointclouds to recognize that you have moved, or do you want to "clean" the images because they are noisy, or something else entirely?
Sorry for not being specific, I edited the post. Please tell me if it's still not clear enough.
This is a good question! Unfortunately I do not know the answer as I am not familiar with such kind of object motion while scanning. Just out of curiosity, can you maybe link the paper that states that the point cloud correction is necessary?
Of course. LiDAR Point Clouds Correction and LiDAR Distortion Correction. Sorry for taking too long to reply, hope it's of any benefit.
Thank you so much!