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

vslam, Running on Your Own Data

asked 2011-08-18 09:55:37 -0600

jrcapriles gravatar image

updated 2011-09-02 09:45:15 -0600

kwc gravatar image

Hi all!

I want to use the vslam package in diamondback for visual odometry. I made all the installation steps, downloaded the stereo tutorial data and tested. It works fine, but know i have some problems trying to use my own data instead of the tutorial bag.

In the section "1.4 Running on Your Own Data" you explain how to use your own data just changing some lines in the stereo_vslam.launch.

My bag had the stereo images unsynchronized, I play my bag and launch the sync_images node and had the stereo images synchronized. But still have some problems using the vslam node. I can run the sync images and the vslam_system without problems, but when i play the bag the vslam_system node says:

[wide_stereo/stereo_vslam_node-1] process has died [pid 20722, exit code -11]. log files: /home/joser/.ros/log/d398d566-cdd9-11e0-8873-f30ac7586bf8/wide_stereo-stereo_vslam_node-1*.log

The only difference that i can see between the playing the two bags (tutorial and my own) using the rxgraph is the subscription of "/tf" topic (with the tutorial bag the vslam_node achieved the subscription and with my bag it didn't)

In the log file it say : [roscpp_internal] [2011-08-24 00:47:55,900] [thread 0x7f53a928c860]: [DEBUG] XML-RPC call [searchParam] returned an error (-1): [Cannot find parameter [tf_prefix] in an upwards search]

edit retag flag offensive close merge delete

Comments

Do you have _exactly_ synchronized images? The tutorial mentions in section 1.2 which messages are required. That was my biggest hurdle to getting started, solved using the ROS node [http://ibotics.ucsd.edu/stingray/browser/branches/rostingray/sync_images | sync_images]. Any error messages for you?
Thomas D gravatar image Thomas D  ( 2011-08-18 11:41:58 -0600 )edit
Please update your original question with new information instead of posting "answers" that don't answer the question. This will help keep ROS Answers easy to read in the future. Thanks for helping to keep ROS Answers well organized!
Eric Perko gravatar image Eric Perko  ( 2011-08-23 11:29:41 -0600 )edit
Thanks you very much!! with the sync_images node I have my stereo images synchronized! But still have some problems with the vslam node.. it says [wide_stereo/stereo_vslam_node-1] process has died [pid 20722, exit code -11]. log files: /home/joser/.ros/log/d398d566-cdd9-11e0-8873-f30ac7586bf8/wide_stereo-stereo_vslam_node-1*.log -
jrcapriles gravatar image jrcapriles  ( 2011-08-24 06:10:40 -0600 )edit

3 Answers

Sort by ยป oldest newest most voted
0

answered 2011-09-02 11:14:54 -0600

Helen gravatar image

vslam is experimental research code. It is not actively being supported, and should be used at your own risk, but patches are welcome.

If you would like a robust method of localization and mapping with a stereo camera or kinect, use the 2D slam_gmapping stack. See the pointcloud_to_laserscan package from the turtlebot stack and the SLAM Gmapping with Kinect Tutorial for the turtlebot.

edit flag offensive delete link more
0

answered 2011-09-04 00:17:40 -0600

davo gravatar image

woops, can get pointcloud from steroe webcams via stereo image proc

I'm just working my way through the topics and trying to set them up....?

Dave

edit flag offensive delete link more
0

answered 2011-09-03 21:31:18 -0600

davo gravatar image

thankyou, for the question and answer been going around in circles for some time on the same issues.

Seems that many are on the same path. I'm having trouble piecing the stages required to gmapping from video.

video, point cloud to laser scan

you need to create point clouds from the video stream, and then convert pointcloud to laserscan message type for gmapping gslam

I know that vslam creates a point cloud, and that eg from the Konnect there is a nodelette that works with a frame throttler to convert point_cloud to laser_scan nodelettes and there seems to be some variations,

In the elektron stack/packages there is a variant launch file with filters for depth (range) etc.

There seems to be a fair amount of remapping and it does not make much sense...

Hope that helps a bit

Dave

edit flag offensive delete link more

Comments

Hi Dave! thanks for your answer, I'm working on visual odometry i tryed with vslam (but like Helen said "vslam is experimental research code") and i already create my own package that create point clouds from the video stream, but then i have problems with the pose estimation (from pcl directly)
jrcapriles gravatar image jrcapriles  ( 2011-09-06 20:43:07 -0600 )edit
Do you know if there is anyway to obtain the pose estimation from pcls or laserscans? or if i use the gmapping gslam (converting my pointcloud to laserscan) is possible to extract the odometry information from it? i do not require the whole SLAM
jrcapriles gravatar image jrcapriles  ( 2011-09-06 20:48:13 -0600 )edit
Laserscan is msg with range start fin and inc angles and time stamp, it has no pose but ref to parent tf frame. u need odom and correct tf tree for range sensor to work with gslam and nav. Not nailed pclouds down yet but expect similar. I was hoping vslam help with poor odometry alas!
davo gravatar image davo  ( 2011-09-13 02:31:57 -0600 )edit
Jose as you have just done a custom conversion could you point me in the right direction, i'm going around in circles as I'm not using a kinect and the tf nodelets etc are jjust confusing me! Dave
davo gravatar image davo  ( 2011-09-13 02:34:33 -0600 )edit
Hi davo, can you tell me a little bit more about your problem? what are you trying to do? do you want to use the gslam node or you are trying to make your own node?
jrcapriles gravatar image jrcapriles  ( 2011-09-14 21:36:10 -0600 )edit
I've hacked the pcl_to_scan for stereo web cam and got scan msgs now need to align range points with minoru fov . not very stable yet pc resource issue I think. using the elektron_kin pkg need to watch the tf's and the filter openni kinect z is height minoru stereo z is depth
davo gravatar image davo  ( 2011-09-15 14:03:09 -0600 )edit

Question Tools

Stats

Asked: 2011-08-18 09:55:37 -0600

Seen: 672 times

Last updated: Sep 04 '11