Visualization with using rosbag
abdullahenesdoruk opened this issue · 14 comments
Hi, how can i visualize with using rosbag file.
eagleye_rt outputs the position of the navsatfix msg /eageye/fix (lat,lon), the position of the custom message /eagleye/enu_absolute_pos_interpolate and the heading angle eagleye/heading_interpolate_3rd.
The package fix2pose in eaglye_util (https://github.com/MapIV/eagleye/tree/master/eagleye_util/fix2pose) publishes geometry_msgs::PoseStamped type message and output tf using these.
So, please run this together
Hi, i ran bag file and eagle_rt launch, after that rosrun fix2kml for visualization. And created kml file. How can i use this file. According to my research, i open google earth and import kml file, is right? Finally, can i visualize like youtube video, if this is possible, how can i do?
According to my research, i open google earth and import kml file, is right?
Right.
Finally, can i visualize like youtube video, if this is possible, how can i do?
To do this demo, you need to use autoware (https://github.com/Autoware-AI/autoware.ai).
By starting up autoware's localization, we output pose by lidar localizar, pcd, and vectormap , pose by eagleye_rt(using fix2pose), and visualize them by rviz .
Data to be used
ROSBAG
https://www.dropbox.com/sh/ks5kg8033f5n3w8/AADv9plEjXnwlxex23Z91kR_a?dl=0
https://www.dropbox.com/sh/mhdib1m1oivotiu/AAD0UnANDsuIsKqcSKHt9WAJa?dl=0
PCD and Vector Map
https://drive.google.com/file/d/1Uwp9vwvcZwaoZi4kdjJaY55-LEXIzSxf/view
I understood, thanks.
Hi, i downloaded rosbag file, i run eagle_rt launch but doesnt work, are rosbag file topics or topics format different?
According to my research, i open google earth and import kml file, is right?
Right.
Finally, can i visualize like youtube video, if this is possible, how can i do?
To do this demo, you need to use autoware (https://github.com/Autoware-AI/autoware.ai). By starting up autoware's localization, we output pose by lidar localizar, pcd, and vectormap , pose by eagleye_rt(using fix2pose), and visualize them by rviz .
Data to be used ROSBAG https://www.dropbox.com/sh/ks5kg8033f5n3w8/AADv9plEjXnwlxex23Z91kR_a?dl=0 https://www.dropbox.com/sh/mhdib1m1oivotiu/AAD0UnANDsuIsKqcSKHt9WAJa?dl=0 PCD and Vector Map https://drive.google.com/file/d/1Uwp9vwvcZwaoZi4kdjJaY55-LEXIzSxf/view
there is no imu or can_twist data in this bag file
You cannot run eagleye with the sample data of autoware ai(sample_moriyama_....bag)..
Please run autoware ai's localizer and eagleye with eagleye's sample data.
the sample data of eagelye
https://www.dropbox.com/sh/ks5kg8033f5n3w8/AADv9plEjXnwlxex23Z91kR_a?dl=0
https://www.dropbox.com/sh/mhdib1m1oivotiu/AAD0UnANDsuIsKqcSKHt9WAJa?dl=0
i got it, downloaded eagle_sample.bag, pcd and vector map.
i select all pcd files for map, csv files for vector map and tf launch file and run it.
and load bag file
and i select eagle_rt.launch file for localization and load.
finally, i open rviz and start bag file
doesnt work. are these steps correct or not?
You cannot run eagleye with the sample data of autoware ai(sample_moriyama_....bag).. Please run autoware ai's localizer and eagleye with eagleye's sample data.
the sample data of eagelye https://www.dropbox.com/sh/ks5kg8033f5n3w8/AADv9plEjXnwlxex23Z91kR_a?dl=0 https://www.dropbox.com/sh/mhdib1m1oivotiu/AAD0UnANDsuIsKqcSKHt9WAJa?dl=0
can you explain? i dont understand.
Please run eagleye_rt.launch in a terminal.
Also, please run fix2poze.launch with it.
And if you add eagleye's pose in the rviz 「add topic」, you can visualize it.
Hi,
i run eagle_rt.launch and then run fixpose. After that open autoware and load pointcloud, vector map and tf.
After that open rviz.
which frame should be selected for target frame and fixed frame. And is neccessary to run lidar localizer?
Sorry for the delay, but I will respond.
You do not need to use lidar localizer.
As for frame selection, my memory is dim, so you can try either one.
Since fix2pose sends tf, I think you can check and debug it.
https://github.com/MapIV/eagleye/blob/master/eagleye_util/fix2pose/src/fix2pose.cpp#L131
This issue is closed for now. If you have problems again, please reopen this issue.