I'm running rtabmap on a jackal robot with bumblebee2 stereo camera on ros-kinetic. I get the following error when trying to view pointcloud2 topic in rviz: [link text](https://imgur.com/a/rOdPHR9). When I view the tf tree [link text](https://imgur.com/a/7e3s1p2) I see that i'm getting -0.086 sec old from the map broadcast by rtabmap.
↧
TF Error: Lookup would requrie extrapolation into the future
↧
Rtabmap PointCloud2 frame oscillates when viewed in RVIZ
The PointCloud2 topic's frame keeps switching back and forth wrt to the robot's frame, as seen in
https://imgur.com/a/NosBsOx. What could be the cause?
↧
↧
RVIZ shows odom frame fixed to map frame when running rtabmap
I've set the map frame as the fixed frame in RVIZ, and this is what it looks like when running rtabmap https://imgur.com/a/U5fJLLP . This is what my partial tf-tree looks like https://imgur.com/a/1s2PnV8. This is my launch file for rtabmap:
↧
Could not get transform from odom to base_link
I'm running rtabmap with an rgb-d sensor and get the following error:
We received odometry message, but we cannot get the corresponding TF odom->base_link at data stamp 1533265358.770186s (odom msg stamp is 1533265358.757750s). Make sure TF of odometry is also published to get more accurate pose estimation. This warning is only printed once.
This is my launch file
↧
Errors using Rtabmap and also in running gmapping and Rtabmap simultaneously.
I have tried many methods suggested in [Kinect + Odometry + 2D laser](http://wiki.ros.org/rtabmap_ros/Tutorials/SetupOnYourRobot) and changed odom and scan topics that suit my robot specifications. But, I could not get any data neither in RViz nor in Rtab GUI. and my launch file is [rtabmap.launch](https://github.com/introlab/rtabmap_ros/blob/master/launch/rtabmap.launch).I am getting this error
[ WARN] [1535559564.824509798]: /rtabmap/rtabmapviz: Did not receive data since 5 seconds! Make sure the input topics are published ("$ rostopic hz my_topic") and the timestamps in their header are set. If topics are coming from different computers, make sure the clocks of the computers are synchronized ("ntpdate"). If topics are not published at the same rate, you could increase "queue_size" parameter (current=10).
/rtabmap/rtabmapviz subscribed to (approx sync):
/rtabmap/odom,
/camera/rgb/image_rect_color,
/camera/depth_registered/image_raw,
/camera/rgb/camera_info,
/rtabmap/odom_info
But when I echoed each topic some data is being published and I am able to get 2D map with this data. But I could not integrate 3D map to this odom and scan data. (odom = IMU and ENCODER data)
I am actually trying to integrate gmapping and RTabmap and found a launch file that suits my requirements. I don't know whether it works. Here is the launch file I am using
↧
↧
working with stereo camera and rtabmap for 3d mapping
So I am using rtabmap and a ZED stereo for this project based o, 3D mapping.
I have the following launch file, I tried to play with with it but no mapping : $(arg initial_pose)
↧
Simultaneous autonomous navigation and mapping
If I understood it correctly, in RTABMap ROS [tutorials](http://wiki.ros.org/rtabmap_ros/Tutorials/MappingAndNavigationOnTurtlebot) mapping and autonomous navigation is a two step process. First mapping is done by driving robot manually and afterwards autonomous navigation can be accomplished using the previously created map.
Is is possible to make my robot autonomously navigate the unknown environment and simultaneously map the area?
I'm running Ubuntu 16.04.4 LTS on an x86 with ROS Kinetic.
Thank you!
↧
rtabmap localization
Hi
I'm trying to run rtabmap in localization mode.
This is my launch file:
filter_field_name: z
filter_limit_min: 0.01
filter_limit_max: 10
filter_limit_negative: False
leaf_size: 0.05
↧
what is output data form of PS 1080 chip used in kinect xbox 360?
please elaborate on data form and also how to process it using ROS?
↧
↧
rtabmap outdoor
When running rtabmap in localization mode (using a map that was made e.g. a few hours before) we're having problems finding loop closures.
I can see that many of the extracted keypoints are features of shadows, and because the sun moves, or it gets clouded temporarily, the loop closures don't happen anymore.
The problem is a lot less when it's clouded.
Is there any way to tune the key point extractor to ignore shadows? Or any other parameters that are useful in these situations?
↧
Parameters for comparing and benchmarking SLAM algorithms
What are the parameters used to compare visual SLAM algorithms?
How do you benchmark and calculate the performance of an algorithm?
I need to evaluate on my own and not use someone else's dataset like [this](https://vision.in.tum.de/data/datasets/rgbd-dataset) one
↧
roslaunch error: Failed to load nodelet
I want to make 2D grid map with RTAB-Map, but I couldn't find the way.
(some people say check "Preperence -> 3D rendering -> Occupancy from 3D cloud projection on the ground", but I can't find it.)
So, I use rtabmap_ros, ros-kinetic with code below.
**roslaunch freencet_launch freenect.launch depth_registration:=true**
**roslaunch rtabmap_ros rgbd-mapping.launch rtabmap_args:="--delete_db_on_start"**
In Ubuntu 16.04, I made it work, but in Windows 10, it made error.
process[camera/points_xyzrgb_hw_registered-16]: started with pid [765] [FATAL] [1510040707.816700900]: Failed to load nodelet '/camera/rgb_rectify_colorof typeimage_proc/rectifyto managercamera_nodelet_manager' [FATAL] [1510040707.850728300]: Failed to load nodelet '/camera/depth_rectify_depthof typeimage_proc/rectifyto managercamera_nodelet_manager' [FATAL] [1510040707.950915300]: Failed to load nodelet '/camera/rgb_rectify_monoof typeimage_proc/rectifyto managercamera_nodelet_manager' [FATAL] [1510040707.951713400]: Failed to load nodelet '/camera/driverof typeopenni_camera/driverto managercamera_nodelet_manager' process[camera/depth_registered_hw_metric_rect-17]: started with pid [775] [FATAL] [1510040708.016043700]: Failed to load nodelet '/camera/depth_metric_rectof typedepth_image_proc/convert_metricto managercamera_nodelet_manager' [FATAL] [1510040708.067463300]: Failed to load nodelet '/camera/ir_rectify_irof typeimage_proc/rectifyto managercamera_nodelet_manager' [FATAL] [1510040708.086891500]: Failed to load nodelet '/camera/rgb_debayerof typeimage_proc/debayerto managercamera_nodelet_manager' process[camera/depth_registered_metric-18]: started with pid [791] process[camera/disparity_depth-19]: started with pid [810] [FATAL] [1510040708.352510500]: Failed to load nodelet '/camera/depth_metricof typedepth_image_proc/convert_metricto managercamera_nodelet_manager'
There's more error logs. If you want, I can add it. How can I make it work?
↧
RTabMap gives only the last scene not the global scene
Through my previous question [here](https://answers.ros.org/question/301960/errors-using-rtabmap-and-also-in-running-gmapping-and-rtabmap-simultaneously/), I am able to run RTabMap without errors. But, all I see in RViz is the last scene not the global scene i.e. I could only see what camera see and the when I move to some other location old data is not visible any further. Any suggestion is appreciated.
Thanks in advance and feel free to comment for any further info needed from my side.
↧
↧
Navigation issue
Hello,
I can't figure out why my real robot does not stop when it arrives close to its goal.
that is my configurations:
I have kinetic distro, ubuntu 16.04
real_sense d435 to make Laserscan topic, encoders for the odometry.
**main launch file:**
↧
rtabmap fails to run a bag of depth and rgdb images
Hi,
I am playing a bag of 16 bits-per-pixel depth and 8 bit rgb images. While playing the bag, I can extract both the rgb and the depth images into files (to verify that the data is ok)
(I used the thread [here](https://answers.ros.org/question/313164/create-a-bag-out-of-16-bit-images/) to create and extract the depth images)
The way the depth image is stored in the bag is by using a python script that uses the PIL package to read a pgm image from file into a PIL object, serialize the object data into byte array (using io.BytesIO()), and save the byte array in PPM format.
To verify the data, I wanted to view the images first, before running rtabmap
In rviz I can view the rgb image, but I cannot view the depth image. According to [here](https://answers.ros.org/question/313323/rviz-not-showing-16-bits-per-pixel-single-channel-depth-images/):
> "RVIZ cannot view 16 bit images
> natively. You'll have to convert them
> to an 8 bit format."
So I assume that the images are correct and move on to use rtabmap
I launch rtabmap using the launch file below:
roslaunch ~/avner/PGR/vision/trunk/ros/catkin_ws/src/bpc_slam_rtabmap/launch/rgbd_slam.rgbd_dataset4.launch
But I cannot see the 3D map in rviz, and I do see error messages from rgbd_odometry
...
BEG getCvType
terminate called after throwing an instance of 'cv_bridge::Exception'
what(): Image is wrongly formed: height * step != size or 385 * 1280 != 492817
Aborted
The depth file is stored the bag in PPM format.
The data payload is 492800 (640*2*385)
- width: 640
- height: 385
- bytes per pixel 2
- step size 640*2 = 1280
The size of the file is 492817, due to the header payload
So actually, the details in the error message are correct ("height*step" should not be equal to "size")
Can someone tell me what am I doing wrong and how to satisfy rgbd_odometry?
Thanks,
Avner
↧
Running rtabmap on bag of rgb and depth images
Hi,
I want to run ros rtabmap on a bag of rgb and depth images, with the following topics:
rostopic list
/camera/depth/camera_info
/camera/depth/image
/camera/rgb/camera_info
/camera/rgb/image_color
I am using the well known canned bag rgbd_dataset_freiburg1_xyz as a reference example. This bag has the following topics:
rosbag info /mnt/hdd3/avnerm/avner/slam/data/rgbd_dataset_freiburg1_xyz/rgbd_dataset_freiburg1_xyz.bag
...
topics: /camera/depth/camera_info
/camera/depth/image
/camera/rgb/camera_info
/camera/rgb/image_color
/cortex_marker_array
/imu
/tf
In my bag I don't have /imu, /tf topics.
To get the program to flow, I added the following transforms.
I added the transform with 0 values for x,y,z,rot,pitch,yaw (i.e. no effect), except for the yaw argument to align the orientation of 3d points with the image
↧
How to clean up a noisy 3d point cloud?
Hi,
I am running ros melodic rtabmap rgbd_odometry. I'm getting a noisy point cloud, which I tried to filter by using the rtabmap_ros parameters:
cloud_noise_filtering_radius, cloud_noise_filtering_min_neighbors
which map to rtabmap parameters
/rtabmap/Grid/NoiseFilteringRadius, /rtabmap/Grid/NoiseFilteringMinNeighbors
I can see that the parameters are passed into the rtabmap process:
ps1 ros | grep -i rtabmap
opt/ros/melodic/lib/rtabmap_ros/rtabmap --udebug --Grid/NoiseFilteringMinNeighbors 80000 --Grid/NoiseFilteringRadius 90000.0 rgb/image:=/camera/rgb/image_color depth/image:=/camera/depth_registered/image_raw rgb/camera_info:=/camera/rgb/camera_info rgbd_image:=rgbd_image_relay left/image_rect:=/stereo_camera/left/image_rect_color right/image_rect:=/stereo_camera/right/image_rect left/camera_info:=/stereo_camera/left/camera_info right/camera_info:=/stereo_camera/right/camera_info scan:=/scan scan_cloud:=/scan_cloud user_data:=/user_data user_data_async:=/user_data_async odom:=odom __name:=rtabmap
I set the parameters to very big values (**NoiseFilteringMinNeighbors 80000 NoiseFilteringRadius 90000.0**), for sanity check.
I expect that the rviz will filter all the 3d points and that the point cloud will be empty, but I'm getting the same point cloud as when cloud_noise_filtering is disabled (NoiseFilteringMinNeighbors 2 NoiseFilteringRadius 0)
rosparam get /rtabmap/Grid/NoiseFilteringMinNeighbors
'2'
rosparam get /rtabmap/Grid/NoiseFilteringRadius
'0'
NoiseFilteringRadius_0__NoiseFilteringMinNeighbors_2
What should I do so that the cloud_noise_filtering takes effect?
Thanks,
Avner
↧
↧
ros rtabmap - graph point cloud vs map point cloud
Hi,
Can someone tell me what is exactly the difference between graph point cloud (/rtabmap/mapData) and map point cloud (/rtabmap/cloud_map)?
As I understand, the map point cloud can be filtered with e.g. the cloud_noise_filtering_radius, cloud_noise_filtering_min_neighbors parameters.
Does this mean that map point cloud is a post-process result of the graph point cloud ?
Is there a way to produce a graph showing the various links in the processing chain?
Thanks,
Avner
↧
setting frame ids for ekf localization
I am using wheel encoders and visual odometry(rtabmap) to perform ekf localization. This is the tf tree of rtabmap in localization mode.
map->odom->camera_link(->other camera frames)
I am trying to inculcate my wheel encoders and use ekf for better localization as rtabmap drifts a lot.
How do i set my frames for wheel encoder odometry publisher and broadcaster?
Also, what goes in the base_link_frame in ekf_localization_node param file?
Thank you
↧
3D map is not created when processing rgbd data using rtabmap with icp_odometry
Hi,
I want to process a bag of rgbd data using rtabmap with icp_odometry.
I can extract a 3d map from a bag that contains rgbd data, using rgbd_odometry.
But when using icp_odometry, the map is not generated...
I can see that the icp_odometry process runs
ps aux ros | grep odom
/opt/ros/melodic/lib/rtabmap_ros/icp_odometry --Reg/Strategy 1 scan:=/scan scan_cloud:=/scan_cloud odom:=odom __name:=icp_odometry __log:=/home/avnerm/.ros/log/0c1e8df6-235c-11e9-b42b-74d02b912ce1/rtabmap-icp_odometry-10.log
I'm seeing some runtime warnings when running icp_odometry:
[ WARN] [1548880246.541540256]: IcpOdometry: Transferring value 5 of "Icp/PointToPlaneK" to ros parameter "scan_normal_k" for convenience.
[ WARN] [1548880246.541801876]: IcpOdometry: Transferring value 1.0 of "Icp/PointToPlaneRadius" to ros parameter "scan_normal_radius" for convenience.
[ WARN] (2019-01-30 12:30:46.542) OdometryF2M.cpp:133::OdometryF2M() OdomF2M/BundleAdjustment=1 cannot be used with registration not done only with images (Reg/Strategy=1), disabling bundle adjustment.
The parameters, and output from the icp_odometry are:
[ INFO] [1548881374.690687725]: Initializing nodelet with 8 worker threads.
[ INFO] [1548881374.977402374, 1548455446.224796923]: Odometry: frame_id = base_link
[ INFO] [1548881374.977433059, 1548455446.224796923]: Odometry: odom_frame_id = odom
[ INFO] [1548881374.977449374, 1548455446.224796923]: Odometry: publish_tf = true
[ INFO] [1548881374.977461946, 1548455446.224796923]: Odometry: wait_for_transform = true
[ INFO] [1548881374.977483511, 1548455446.224796923]: Odometry: wait_for_transform_duration = 0.100000
[ INFO] [1548881374.977521110, 1548455446.224796923]: Odometry: initial_pose = xyz=0.000000,0.000000,0.000000 rpy=0.000000,-0.000000,0.000000
[ INFO] [1548881374.977538825, 1548455446.224796923]: Odometry: ground_truth_frame_id =
[ INFO] [1548881374.977554142, 1548455446.224796923]: Odometry: ground_truth_base_frame_id = base_link
[ INFO] [1548881374.977570620, 1548455446.224796923]: Odometry: config_path =
[ INFO] [1548881374.977585390, 1548455446.224796923]: Odometry: publish_null_when_lost = true
[ INFO] [1548881374.977599455, 1548455446.224796923]: Odometry: guess_frame_id =
[ INFO] [1548881374.977614073, 1548455446.224796923]: Odometry: guess_min_translation = 0.000000
[ INFO] [1548881374.977629616, 1548455446.224796923]: Odometry: guess_min_rotation = 0.000000
[ INFO] (2019-01-30 12:49:35.080) Parameters.cpp:907::parseArguments() Parsed parameter "Reg/Strategy"="1"
[ INFO] [1548881375.080094503, 1548455446.224796923]: Update odometry parameter "Reg/Strategy"="1" from arguments
[ WARN] [1548881375.102105584, 1548455446.224796923]: IcpOdometry: Transferring value 5 of "Icp/PointToPlaneK" to ros parameter "scan_normal_k" for convenience.
[ WARN] [1548881375.102373097, 1548455446.224796923]: IcpOdometry: Transferring value 1.0 of "Icp/PointToPlaneRadius" to ros parameter "scan_normal_radius" for convenience.
[DEBUG] (2019-01-30 12:49:35.102) Odometry.cpp:63::create() type=0
[DEBUG] (2019-01-30 12:49:35.102) OdometryF2M.cpp:77::OdometryF2M()
[DEBUG] (2019-01-30 12:49:35.102) Registration.cpp:48::create() type=1
[ WARN] (2019-01-30 12:49:35.102) OdometryF2M.cpp:133::OdometryF2M() OdomF2M/BundleAdjustment=1 cannot be used with registration not done only with images (Reg/Strategy=1), disabling bundle adjustment.
[ INFO] [1548881375.106376949, 1548455446.224796923]: IcpOdometry: scan_cloud_max_points = 0
[ INFO] [1548881375.106407441, 1548455446.224796923]: IcpOdometry: scan_downsampling_step = 1
[ INFO] [1548881375.106449822, 1548455446.224796923]: IcpOdometry: scan_voxel_size = 0.000000
[ INFO] [1548881375.106461144, 1548455446.224796923]: IcpOdometry: scan_normal_k = 5
[ INFO] [1548881375.106471202, 1548455446.224796923]: IcpOdometry: scan_normal_radius = 1.000000
Here is the launch file:
↧