Taiga

@YoshiRi The PR has been revised to address certain points that were raised, and is now in a state where it can be merged. As for the changes made to...

I measured the latency in the subscriber function onOccupancyGridMapAndPointCloud2 and mainly improved the latency in splitPointCloudFrontBack. The changes made the other day have not improved the speed as initially anticipated....

@veqcc When comparing with percentiles, there is a slight improvement. | PERCENTILE | Before [ms] | After [ms] | |-------------|-------------|------------| | 99% | 16.9 | 16.4 | | 90% |...

@veqcc First, it should be noted that the `splitPoint...` function is called within the `onOccupancy...` function. And since the `onOccupancy...` function is the main callback for this node, it measures...

@veqcc I'm sorry. I misunderstood the graph. Tomorrow, I will check if the numbers are incorrect. It seems that I might have confused "before" and "after."

@veqcc As a result of measuring performance to verify once again based on the feedback, performance improvements were observed in both the `splitPoint...` function and the `onOccupancy...` function.  **onOccupancyGridMapAndPointCloud2**...