I have a car with 8 lidars, each with a field of view of 60 degrees. My car looks like this:


How can I merge all the lidar readings into 1 point cloud?

  • $\begingroup$ I think SignalProcessing.SE can help you with this problem better. $\endgroup$ – DuttaA Oct 13 '19 at 15:36
  • $\begingroup$ ... with the help of sensor fusion. The idea is combine separate uncertain data input into a reliable environment picture. In the easiest case the result is a panoramic image and the next question is perhaps how to detect objects in this mosaic... $\endgroup$ – Manuel Rodriguez Oct 13 '19 at 21:53
  • $\begingroup$ Hi Josh. I think you should clarify how this problem is related to AI. $\endgroup$ – nbro Oct 20 '19 at 15:11

A high-level solution is to use the NVIDIA DriveWorks SDK. NVIDIA's Developer Blog has a good post, "Point Cloud Processing with NVIDIA DriveWorks SDK".

The NVIDIA DriveWorks SDK contains a collection of CUDA-based low level point cloud processing modules optimized for NVIDIA DRIVE AGX platforms. The DriveWorks Point Cloud Processing modules include common algorithms that any AV developer working with point cloud representations would need, such as accumulation and registration.

I believe you don't need the AGX hardware for the problem you posted.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.