Bugs about euclidean_cluster

[beginningfan]

Checklist

• I've read the contribution guidelines.
• I've searched other issues and no duplicate issues were found.
• I'm convinced that this is not my fault but a bug.

Description

1. min_cluster_size is not passed to clustering algorithm
2. When the max_cluster_size_ is large, clustering takes too much CPU and memory because this line. This bug is related to this pr

Expected behavior

1. The size of the cluster is greater than or equal to the min_cluster_size
2. Memory allocation should be based on the actual size of the point cloud

Actual behavior

1. The input parameter of the cluster algorithm is 1
2. Memory allocation is based on the maximum number of clusters

Steps to reproduce

1. print size of clusters before filter in here
2. use a large number such as 30000 as max_cluster_size in config

Versions

OS: ROS2
autoware: latest

Possible causes

No response

Additional context

No response

[badai-nguyen]

@beginningfan thank you for your raising issue,
Regarding to your 1.

min_cluster_size is not passed to clustering algorithm

I think that min_cluster_size was passed and used to check clusters' size here:

autoware.universe/perception/euclidean_cluster/lib/voxel_grid_based_euclidean_cluster.cpp

Lines 134 to 136 in b284599

	if (!(min_cluster_size_ <= static_cast<int>(i_cluster_data_size / point_step) &&
	static_cast<int>(i_cluster_data_size / point_step) <= max_cluster_size_)) {
	continue;

Could check it again?