search

erik-nelson / blam

Other
790 stars 348 forks source link

blam_slam_node run error #21

Open manomitbal opened 7 years ago

manomitbal commented 7 years ago

Hi!

I would like to start by congratulating you on a lovely algorithm. That being said, I am encountering a very peculiar error where the transformation matrices are giving out NaN values whenever I try to launch the blam slam node. The following is the output:

`started roslaunch server http://192.168.1.77:43921/

SUMMARY

PARAMETERS

NODES /blam/ blam_slam (blam_slam/blam_slam_node)

ROS_MASTER_URI=http://192.168.1.80:11311

core service [/rosout] found process[blam/blam_slam-1]: started with pid [4733] [ INFO] [1498763198.239834200]: /blam/blam_slam/BlamSlam: Registering online callbacks. blam_slam_node: /build/pcl-6_P28C/pcl-1.7.2/kdtree/include/pcl/kdtree/impl/kdtree_flann.hpp:136: int pcl::KdTreeFLANN<PointT, Dist>::nearestKSearch(const PointT&, int, std::vector&, std::vector&) const [with PointT = pcl::PointXYZ; Dist = flann::L2_Simple]: Assertion point_representation_->isValid (point) && "Invalid (NaN, Inf) point coordinates given to nearestKSearch!"' failed. [blam/blam_slam-1] process has died [pid 4733, exit code -6, cmd /home/fx250/blam_ws/devel/lib/blam_slam/blam_slam_node ~pcld:=/zed/point_cloud/cloud_registered __name:=blam_slam __log:=/home/fx250/.ros/log/166c611c-5cfb-11e7-8538-001fc69c3cb2/blam-blam_slam-1.log]. log file: /home/fx250/.ros/log/166c611c-5cfb-11e7-8538-001fc69c3cb2/blam-blam_slam-1*.log all processes on machine have died, roslaunch will exit shutting down processing monitor... ... shutting down processing monitor complete done

Any help is appreciated! Thanks in advance!

Nickhodem commented 7 years ago

Hi, I think you should remove every NaN points in point cloud, before you sent it to blam. pcl::PointCloud laserCloudIn; pcl::PointCloud::Ptr laserCloudWithoutNaN (new pcl::PointCloud ()); std::vector indices; pcl::removeNaNFromPointCloud(laserCloudIn, *laserCloudWithoutNaN, indices);

tiberium24 commented 5 years ago

These directions are not correct. Can someone say where we should make this conversion and how? Since it is very common to get Nan values, we should fix this.

Disonye commented 5 years ago

filter里面没有去除nan点，建议加载点云后去除nan点； if (!points->is_dense) { points_filtered->is_dense = false; std::vector indices; pcl::removeNaNFromPointCloud(points_filtered,points_filtered, indices); }

shrr98 commented 4 years ago

filter里面没有去除nan点，建议加载点云后去除nan点； if (!points->is_dense) { points_filtered->is_dense = false; std::vector indices; pcl::removeNaNFromPointCloud(points_filtered,points_filtered, indices); }

does this really fix the problem? I have tried the solution but the issue is still there.