Ouster LiDAR Performance in ROS2

This package provides an evolving set of tools to help quantify the performance of using Ouster LiDARs within ROS2. We have several objectives in developing this framework and set of tools. They are:

1. To quantitatively understand the performance characteristics that can be expected when using Ouster LiDARs in ROS2
2. To identify any performance bottlenecks with the intention of giving a directed way to address the issues
3. To work toward an optimized ROS2 Ouster LiDAR technology stack

Performance Analysis

On-going

1. Foxy Performance Analysis: Jitter and end-to-end latency analysis is being conducted against ROS2 Foxy for various LiDAR modes. This work builds upon the extensive foundational analysis conducted against ROS2 Eloquent.

Completed

1. Eloquent Performance Analysis: Significant work was conducted to get a baseline sense of the jitter and latency we can expect when using Ouster LiDARs with ROS2. This was done in ROS2 eloquent.
2. Data organization: In this PR conversation, it was alluded to that some performance hit is being taken by how the LiDAR data are organized in the memory buffers of the driver. This notebook articulates exactly how those data are laid out in memory and proposes a (simple) reshaping filter that could be applied in the driver to save all consuming applications from having to do this. The hope is that it will lead to better overall performance and provide a more inuitive interface to the data. This notebook serves as the motivation for some work that will be conducted in the driver code itself. UPDATE: As a result of this work, the driver has been updated in this PR to reflect the suggested data layout changes to the Point Cloud.