ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
0

Slam gmapping tutorial troubleshooting

asked 2011-07-19 14:31:41 -0500

Khiya gravatar image

Hello, I'm following the map-building tutorial at http://www.ros.org/wiki/slam_gmapping... and I can get all the way through step 2 (hurray). Then when I enter:

 $ rosbag play aptLaserData.bag

which is the second part 1 of step 2, I get the error:

[ INFO] [1311128237.384522694]: Opening aptLaserData.bag

[FATAL] [1311128237.388682646]: Time is out of dual 32-bit range

My .bag file is indeed named aptLaserData.bag, but could you tell me what's causing the problem?

Thanks, Khiya

edit retag flag offensive close merge delete

Comments

Another user reported the same problem, using the bag supplied with the tutorial: http://answers.ros.org/question/248/slam_gmapping-mappingfromloggeddata-out-of-dual-32. I wasn't able to reproduce it. Can you post your bag somewhere that we can try playing it back?
Brian Gerkey gravatar image Brian Gerkey  ( 2011-07-20 03:42:18 -0500 )edit
Brian, I can't find a way to upload a .bag file onto ROS answers. Unless you know of a better way to put my data online, I'm going to upload it to the repository for one of my other projects and make it available there.
Khiya gravatar image Khiya  ( 2011-07-20 06:10:44 -0500 )edit
PS, could this problem have to do with the fact that I'm using a virtual machine running Ubuntu 10.10 from a 32-bit Windows machine running Vista?
Khiya gravatar image Khiya  ( 2011-07-21 03:01:18 -0500 )edit
Hi Brian, thanks for your patience! Please see my comment on Tim's post, below, for access to my aptLaserData.bag file.
Khiya gravatar image Khiya  ( 2011-07-28 05:39:48 -0500 )edit

5 Answers

Sort by ยป oldest newest most voted
0

answered 2011-09-02 10:56:44 -0500

kwc gravatar image

No activity in a month, closing. In the future, please use the 'edit' feature to update the original question, or 'post a comment' feature in order to do followup so that it is more easy to track the state of the Q&A.

edit flag offensive delete link more
0

answered 2011-07-20 06:01:41 -0500

Khiya gravatar image

One more thing: when I downloaded the provided .bag file from Willow Garage and tried playing it, this is what happened:

    $ rosbag play basic_localization_stage.bag
[ INFO] [1311183994.765718223]: Opening basic_localization_stage.bag

Waiting 0.2 seconds after advertising topics... done.

 Hit space to toggle paused, or 's' to step.
 [RUNNING]  Bag Time:     34.600000   Duration: 0.000000 / 1239670987.745222    
 [RUNNING]  Bag Time:     34.600420   Duration: 0.000420 / 1239670987.745222   
 [RUNNING]  Bag Time:     34.703970   Duration: 0.103970 / 1239670987.745222    
 [RUNNING]  Bag Time:     34.872898   Duration: 0.272898 / 1239670987.745222

...etc. It ran until bag time hit 126, then this message came up:

 [RUNNING]  Bag Time:    126.300712   Duration: 91.700712 / 1239670987.745222   
 [RUNNING]  Bag Time:    126.400910   Duration: 91.800910 / 1239670987.745222   
 [RUNNING]  Bag Time:    126.510560   Duration: 91.910560 / 1239670987.745222    
 [RUNNING]  Bag Time:    126.600720   Duration: 92.000720 / 1239670987.745222  
 [FATAL] [1311182318.849101157]: Time is out of dual 32-bit range

I went ahead with the next few steps in the tutorial anyway, and generated a partial map off the data from WG, below. Partial map from WG laser data

(not sure what's going on with the picture there - I'll try to get it up somehow)

Any enlightenment?

Thanks,

-Khiya

edit flag offensive delete link more
0

answered 2011-07-19 19:22:13 -0500

jan gravatar image

Hi Khiya,

I am very new at ROS.org but I had a similar problem. First of all did you set your time to sim_time?

$ rosparam set use_sim_time true

After that try this command:

$ rosbag play aptLaserData.bag --clock

If it works let me know.

Jan

edit flag offensive delete link more

Comments

Hi Jan, thanks for the suggestion, but it didn't work. I also tried setting use_sim_time to false, just in case, but of course that didn't work either. I'm still getting the exact same error either way.
Khiya gravatar image Khiya  ( 2011-07-20 02:53:02 -0500 )edit
I also forced use_sim_time to be false while I was logging the data, then set it back to true while trying to play it, but still no luck.
Khiya gravatar image Khiya  ( 2011-07-28 09:01:23 -0500 )edit
When should I use parameter --clock?When should not? Thank you~
sam gravatar image sam  ( 2011-07-28 13:21:54 -0500 )edit
Hm I dont know. Normally I did it every time when I let a rosbag-file play. @ Khiya I tried to reproduce the error but it works on my system without a problem. Have you tried to install a normal Ubuntu not in a VM?
jan gravatar image jan  ( 2011-07-28 18:51:50 -0500 )edit
Hi Jan, yes, I did test the same laser on an Ubuntu machine, same results. Sad Trombone.
Khiya gravatar image Khiya  ( 2011-07-29 01:36:19 -0500 )edit
0

answered 2011-07-27 06:23:46 -0500

Tim Field gravatar image

Khiya, I'd like to reproduce your problem. Can you please give me a link to aptLaserData.bag?

Also, I reindexed the basic_localization_stage.bag referenced in the tutorial. If you have a chance, can you try downloading the bag again and see if you have the same problem.

edit flag offensive delete link more

Comments

Tim, thanks for your patience. I've added aptLaserData.bag to my current project trunk at googlecode, https://e-bot.googlecode.com/svn/trunk/ . I think you should be able to use svn to get a copy. And if you know of a better way to pass .bag files around the Internet, please let me know. Thanks!
Khiya gravatar image Khiya  ( 2011-07-28 05:38:22 -0500 )edit
Tim, I also just tried reindexing basic_localization_stage.bag before playing it. Unfortunately the same error occurred at almost the exact same time.
Khiya gravatar image Khiya  ( 2011-07-28 06:01:45 -0500 )edit
Just in case it's relevant, I first calibrated my laser scanner following the basic suggestions at http://www.ros.org/wiki/hokuyo_node/Tutorials/UsingTheHokuyoNode . I started hokuyo_node, then began the gmapping tutorial above. Could it just be a configuration I've missed?
Khiya gravatar image Khiya  ( 2011-08-02 07:48:47 -0500 )edit
0

answered 2011-08-03 09:10:35 -0500

Khiya gravatar image

I've found a partial solution: I do not have a fully functional robot yet - I was assuming I could just set up slam_gmapping with a Hokuyo URG-04LX laser ranger and an ASUS EeeBox running ROS. That means I had not set up my /tf or /odom nodes. I'm not sure why this caused trouble, but apparently it did. As a temporary fix, I used this launch file to initialize the laser before I ran through the steps in slam_gmapping tutorial:

 <launch>
    <node pkg="hokuyo_node" type="hokuyo_node" name="scan">
    <param name="port" value = "/dev/robot/laser" />
    <param name="max_ang_degrees" value="90" />
    <param name="min_ang_degrees" value="-90" />
    <param name="frame_id" value="base_scan" />
    </node>

    <node pkg="tf" type="static_transform_publisher" name="hokuyo_static_tf_publisher" args="0 0 0 0 0 0 1 base_link base_scan 50" />

    <node pkg="tf" type="static_transform_publisher" name="odom_static_tf_publisher" args="0 0 0 0 0 0 1 odom base_link 50"/>

</launch>

All I did in this file was set up a static identity transform from base_link to base_scan, and from odom to base_link. The dual 32-bit error did not appear, thankfully, but another problem did happen: In the terminal running slam_gmapping, this warning appeared when I played back the bag file:

[WARN] [1312405487.766138148]: TF_OLD_DATA ignoring data from the past for frame /odom at time 1.31241e+09 according to authority /play_1312405475737786833
Possible reasons are listed at 
[ WARN] [1312405487.768969562]: TF_OLD_DATA ignoring data from the past for frame /base_scan at time 1.31241e+09 according to authority /play_1312405475737786833
Possible reasons are listed at 
[ WARN] [1312405487.780878676]: TF_OLD_DATA ignoring data from the past for frame /odom at time 1.31241e+09 according to authority /play_1312405475737786833
Possible reasons are listed at

and then nothing, so I don't know where 'Possible reasons' are listed. After that, when I ran the map_saver function, it began but I received a 'Waiting for map' message that did not end until I canceled the process.

There must be something else that's obvious, I'm just missing it. Can anyone spot it?

Thanks,

Khiya

edit flag offensive delete link more

Question Tools

Stats

Asked: 2011-07-19 14:31:41 -0500

Seen: 2,177 times

Last updated: Sep 02 '11