search

Shawn207 / map_manager_pub

3D Dynamic Map for robot: A real-time dynamic obstacle tracking and mapping system for UAV navigation and collision avoidance with an RGB-D camera
45 stars 8 forks source link

rostopic data output error #5

Open yangzhongii opened 2 weeks ago

yangzhongii commented 2 weeks ago

Thank you to the author for opening up such a perfect job. However, I encountered an error when calling the corresponding file according to your tutorial document. According to the document, replace the corresponding topic with the one on my robot. But when I replaced it, the map data corresponding to occupency_map could not be displayed on rviz, and the terminal could not display any errors. Afterwards, I made modifications according to the camera's display size, which were consistent with yours, but still couldn't be displayed. When I used rostopic echo to read the corresponding point cloud data, I found that the corresponding data was inconsistent with yours. It repeatedly displays the header: seq: 120 stamp: secs: 0 nsecs: 0 frame_id: "odom" height: 1 width: 0 fields: name: "x" offset: 0 datatype: 7 count: 1 name: "y" offset: 4 datatype: 7 count: 1 name: "z" offset: 8 datatype: 7 count: 1 is_bigendian: False point_step: 16 row_step: 0 data: [] is_dense: True And the data has not changed. The data format generated by my three topics is consistent with yours. I didn't know there was a problem there. I hope you can help me answer this question and answer in your spare time. I would greatly appreciate it.

yangzhongii commented 2 weeks ago

And here is my rqt_graph rqt_graph

Shawn207 commented 1 week ago

Sorry for the late reply. It seems that you are using depth image as sensor input. In this case, we set sensor_input_mode as 0, and subscribe the depth_image instead of corresponding pointcloud. We did not test pointcloud from sensor image, so I guess that might be the reason you are getting inconsistency. You can try accept depth_image instead. also, since you are using /odom, make sure the localization_mode is 1

ltf1001 commented 1 week ago

Hello author, I also encountered the same problem. I replaced the robot with a ground mobile robot. And replaced the topic (according to the readme and the results of the previous question). The corresponding map information results cannot be displayed in rviz. Do I need to improve other places to display the corresponding map information?

Shawn207 commented 4 days ago

Hello author, I also encountered the same problem. I replaced the robot with a ground mobile robot. And replaced the topic (according to the readme and the results of the previous question). The corresponding map information results cannot be displayed in rviz. Do I need to improve other places to display the corresponding map information?

Did you tried the suggestions above?

ltf1001 commented 4 days ago

作者你好，我也遇到了同样的问题，我将机器人换成了地面移动机器人，并且替换主题（按照readme和上一个问题的结果），rviz中仍然无法显示相应的地图信息，我是否需要改进代理才能显示相应的地图信息呢？

您尝试过上述建议吗？

Thanks for the author's reply. I have followed your suggestions and made the changes as above. But it still doesn't solve my problem. I followed your configuration and matched the information sent by the corresponding depth camera and the corresponding coordinates with mine. But the information sent by the corresponding topic does not match yours at all, and the output is continuous 1. There is no error in the result, I don't know how to modify it. I really hope to get your help.

Shawn207 commented 4 days ago

In this case, could you check your time stamp of the odom and depth info? We use ros time synchronizer for mapping, so that requires both. If that does not work, could you send your output error log? BTW, what do you mean "Out put is continuous 1"? what output are you talking about?