|
earth_rover_localization package from earth_rover_localization repoearth_rover_bed_detection earth_rover_localization |
|
|
Package Summary
| Tags | No category tags. |
| Version | 1.2.0 |
| License | BSD |
| Build type | CATKIN |
| Use | RECOMMENDED |
Repository Summary
| Checkout URI | https://github.com/earthrover/earth_rover_localization.git |
| VCS Type | git |
| VCS Version | master |
| Last Updated | 2022-08-02 |
| Dev Status | MAINTAINED |
| CI status | Continuous Integration |
| Released | RELEASED |
| Tags | No category tags. |
| Contributing |
Help Wanted (0)
Good First Issues (0) Pull Requests to Review (0) |
Package Description
Additional Links
Maintainers
- Andres Palomino
- Ricard Pardell
Authors
Earth Rover localization
This package contains ROS nodes, configuration and launch files to use the EKF of the robot_localization package with the Earth Rover Open Agribot. The package has been tested in Ubuntu 16.04.3 and ROS Kinetic. If you don't have ROS installed, use the following line.
$ sudo apt-get install ros-kinetic-ros-base
Dependencies
A summary description and links to the corresponding sites are listed below only if browsing further information is needed. The installation steps will download or include necessary configuration details on how to use the required dependences
robot_localization: Robot Localization is a collection of state estimation nodes, each of which is an implementation of a nonlinear state estimator for robots moving in 3D space. It contains two state estimation nodes, ekf_localization_node and ukf_localization_node. In addition, robot_localization provides navsat_transform_node, which aids in the integration of GPS data. A more detailed documentation explaining this package (nodes, workflow, ...) can be found here.
Piksy Driver: Driver compatible with the Swift Navigation devices.
Xsense Driver: Driver for Xsens IMU devices.
GeographicLib: Offers C++ interfaces to a set of geographic transformations.
Dependency on host computer to monitor results.
- Mapviz: Visualization tool with a plug-in system similar to RVIZ focused on visualizing 2D data.
Installation and Configuration
- On the on-board PC, create a workspace and clone the repository.
$ mkdir -p ~/earth_rover_ws/src
$ cd ~/earth_rover_ws/src
$ git clone --recursive https://github.com/earthrover/earth_rover_localization.git
GeographicLib
- Download GeographicLib
$ cd ~/earth_rover_ws
$ wget https://sourceforge.net/projects/geographiclib/files/distrib/GeographicLib-1.49.tar.gz/download
- Installation using CMAKE
- Unpack the source
tar xfpz download
- then enter the directory created, create a separate build directory and enter it.
cd GeographicLib-1.49
mkdir BUILD
cd BUILD
- Run cmake, pointing it to the source directory (..)
cmake ..
- Build and install the software. if CMAKE_INSTALL_PREFIX is a system directory
make
sudo make install
Further installing details can be found [here](https://geographiclib.sourceforge.io/html/install.html).
Robot_localization and Piksy RTK packages
- Install the robot localization package
sudo apt-get install ros-$ROS_DISTRO-robot-localization
**NOTE**: We currently use version **2.6.9**. If the installation with apt installs a newer version, follow the next steps:
- Delete the source installation:
sudo apt-get remove ros-melodic-robot-localization
- Clone the repo:
cd ~/earth_rover_ws/src/libs
git clone https://github.com/cra-ros-pkg/robot_localization.git
- Checkout the corresponding tag version:
cd ~/earth_rover_ws/src/libs/robot_localization
git checkout tags/2.6.9
- Download ROS drivers for the Piksi RTK GPS module
mkdir -p ~/earth_rover_ws/src/libs
cd ~/earth_rover_ws/src/libs
git clone https://github.com/earthrover/earth_rover_piksi
- Checkout the new feature which isn't in the release yet (To test)
cd ~/earth_rover_ws/src/libs/earth_rover_piksi
git checkout feature_2.14.1
Xsense
- Download the ROS driver for Xsens MTi devices
$ cd ~/earth_rover_ws/src/libs/
$ git clone https://github.com/earthrover/earth_rover_xsens.git
- Build xspublic from your catkin workspace:
$ pushd src/xsens_ros_mti_driver/lib/xspublic && make && popd
Compile
- Compile the packages
$ cd ~/earth_rover_ws
$ catkin_make
$ source devel/setup.bash
- Drivers configuration to be performed on the embedded device.
- Configure the launch files for the package on each receiver (Reference and Attitude) according to the desired IP addresses. The configuration files can be found in
earth_rover_localization/launch/piksi_multi_rover_reference.launchandearth_rover_localization/launch/piksi_multi_rover_attitude.launchrespectively. This repository includes by defaultreference receiver ip_address to 192.168.0.222andattitude receiver ip_address to 192.168.0.223
Check the <arg name="interface" is set to default="tcp" and change the desired tcp_addr if necessary.
<!-- If interface is tcp, this specifies the address of Piksi Multi. -->
<arg name="tcp_addr" default="192.168.0.222"/>
- The ROS node reads SBP (Swift Navigation Binary Protocol) messages, a fast, simple, and minimal overhead binary protocol for communicating with Swift Navigation devices.
Please check here which Piksi Multi firmware version based on the current SBP Lib version.
Currently the scripts/install_sbp.sh will install SBP Lib 2.4.1.
This means you are supposed to install Firmware 2.1.14 from SwiftNav Firmware page in your Piksi Multi.
WARNING: If upgrading from a firmware below v2.0.0 to a firmware greater than v2.0.0, you must upgrade to v2.0.0 first.
WARNING: install ONLY ONE version of SBP library, depending of which Hardware version you are using.
The following code will automatically download the required version of libsbp and install it in the default folder /usr/local/lib/python2.7/dist-packages/sbp-<SBP_LIB_VERSION>-py2.7.egg/sbp/.
# To install SBP, Execute this line in the package folder 'earth_rover_localization'
source scripts/install_sbp.sh
Check that the installations finishes succesfully. If any error appears due to locale settings, check the following solution and try the SBP installation again.
- To configure the ENU results from the ROS driver, fill the
enu_origin.yamlon the package folderearth_rover_localization/cfgwith the same location of the base station. If base station is not yet configured, remember to edit this file before launchingearth_rover_localization
Hardware configuration
The following steps explain the Hardware and ROS drivers configuration to run in the embedded device and monitor on a host PC. Only follow this section if the required sensors are available to test for a complete base station - rover setup. Skip to Earth Rover localization to see robot localization node, visualization tool and play recorded results.
Piksy Modules
The Hardware configuration uses three Piksi Multi Evaluation Board: A base station and two receivers on the rover (reference and attitude).
The reference receiver obtains corrections from base station using the FreeWave Radio Evaluation Kit and then send corrections to the attitude receiver which enables precise heading output.
- Swift console on a host computer
- Follow the instructions on installing and operating the Swift Console on your computer, see the Installation Instructions and User's Guide.
- Use the following steps to configure the Piksy modules.
Complete the instructions to configure the base station and rover receiver to use the GNSS RTK Position with Stationary Base solution.
Follow the configuration to enable the heading setup. Use the piksis on the rover to configure this step. WARNING Be aware that one receiver (reference receiver) has already be configured to receive corrections from a base station. Configure the
enabled_sbp_messageson uart1 instead to send corrections from reference to attitude through this available port. See the documentation details.
- Enable the Ethernet Configuration on reference and attitude receivers based on the desired IP addresses to connect.
Usage
- Test the ROS node on the reference receiver. The following line will launch the configuration over TCP/IP. Check that observations are received on the Swift Console and also the published topics from the driver
$ roslaunch earth_rover_localization piksi_multi_rover_reference.launch
- Test the ROS node on the attitude receiver. Check again observations and topics specially the
/piksi_attitude/baseline_heading
$ roslaunch earth_rover_localization piksi_multi_rover_attitude.launch
- Test the xsense node and check the publised topics.
$ roslaunch earth_rover_localization xsens.launch
- The complete launch can be used to include sensor drivers and robot localization node to estimate the robot's pose in real time.
$ roslaunch earth_rover_localization er_localization_rover.launch
Earth Rover localization
Robot localization package
If not already installed, install the robot localization package
$ sudo apt-get install ros-$ROS_DISTRO-robot-localization
The package includes .bag example files of recorded tracks to run robot localization and tune the EKF params if necessary.
The following launch file reproduces a bag file and applies the robot localization to adquire the pose estimation of the rover.
$ roslaunch er_localization er_localization_player.launch
The result of the localization package is the robot's pose estimation in its world frame. Then, the origin of the world frame is georeferenced and will change depending on where the scouting mission is performed.
The launch will find the coordinates of the base station to set the origin of the Map frame. Check the enu_origin.yaml on the package folder earth_rover_localization/cfg. Only if you're using the bag player, set the following location on the configuration file.
latitude0_deg: 41.4678702
longitude0_deg: 2.0227646
altitude0: 133.0527
Inputs
The main published topics are:
- /imu/data: A sensor_msgs/Imu.msg with the imu data
- /heading: A sensor_msgs/Imu.msg which reports the vehicle heading reported by the baseline between reference and attitude receiver
- /piksi_receiver/navsatfix_best_fix: A sensor_msgs/NavSatFix.msg that contains WGS 84 coordinates with best available fix at the moment (either RTK or SBAS)
Nodes
Used nodes on the architecture.
- navsat_transform: Takes as input the GPS data and produces an odometry message in coordinates that are consistent with the robot’s world frame. More info about navsat_transform_node on Documentation.
- ekf_localization: The node is an implementation of an Extended Kalman filter. It uses an omnidirectional motion model to project the state forward in time, and corrects that projected estimate using perceived sensor data. Detailed information on the Documentation page.
- heading_listener: In charge of traducing the vehicle heading from the baseline into ROS-compliant message to use in robot localization node
- set_datum: Finds the convergence value of the desired start location in enu_origin.yaml. It's necessary to set the map's origin.
- remove_latency: This node corrects the latency of the GPS/EKF by subtracting a constant value to the message timestamps, passed as a parameter.
- set_initial_state_filter: Takes a GPS measurement and transforms it into odom frame to set the initial_state parameter for the EKF filter in the robot_localization pkg. Waits until RTK fix is achieved to use the GPS data and save the initial state parameter.
Outputs
-
/odometry/gps: A nav_msgs/Odometry.msg message containing the GPS coordinates of your robot, transformed into its world coordinate frame. -
gps/filtered: A sensor_msgs/NavSatFix.msg message containing your robot’s world frame position, transformed into GPS coordinates. -
/odometry/filtered/global: A nav_msgs/Odometry.msg message with the robot's pose estimation in its world frame. -
/odometry/filtered/global/corrected_latency: A nav_msgs/Odometry.msg message with a timestamp corrected with a specified latency. Requires running theremove_latencynode.
Mapviz (Optional) on host PC
The idea of this step is to start a ROS system using two machines and monitorize the master on the embedded PC from a host PC. Host PC and embedded PC should be connected on the same network. Get involve with working with multiple machines and the environmental variable ROS_MASTER_URI here
You can install mapviz using apt-get from the ROS apt repository:
$ sudo apt-get install ros-$ROS_DISTRO-mapviz ros-$ROS_DISTRO-mapviz-plugins ros-$ROS_DISTRO-tile-map ros-$ROS_DISTRO-multires-image
Go to this tutorial to enable ROS Offline Google Maps for MapViz
- Go to the downlodaded package and edit the launch file
$ roscd mapviz
$ cd launch/
$ sudo nano mapviz.launch
- Replace the file with the following template for
mapviz.launchEdit the local_xy_origins with the base station location
<?xml version="1.0"?>
<launch>
<!-- Mapviz -->
<arg name="print_profile_data" default="true"/>
<node pkg="swri_transform_util" type="initialize_origin.py" name="initialize_origin" output="screen">
<param name="local_xy_frame" value="map"/>
<param name="local_xy_origin" value="swri"/>
<rosparam param="local_xy_origins"> # Replace with Map origins. Same as robot localization datum param
[{ name: swri,
latitude: 41.4678702,
longitude: 2.0227646,
altitude: 132.530939,
heading: 0}]
</rosparam>
</node>
<node pkg="mapviz" type="mapviz" name="mapviz"></node>
</launch>
- launch mapviz
$ roslaunch mapviz mapviz.launch
Download the config file from this repository,
cfg/mapviz_localization_config.mvc, go toFile -> Open Configon the top bar and upload the configuration file.Visualization of topics should appear to follow the localization task.
Documentation
The documentation provided above gives general knowledge on how to install, configure and use the package in a general way. In this section, an in depth explanation of the package is provided. A knowledge of the robot localization package is required.
TF tree
In order to understand a little better the robot model and it's references in the next sections, see the full Tf tree structure in the documentation for the ugv description package.
The localization package provides the transforms between map, odom, utm and rover_base_link frames.
The robot localization package needs to be configured in a very specific way to build properly the TF tree:
-
LOCAL EKF: Run one instance of a
robot_localizationstate estimation node that fuses only continuous data, such as odometry and IMU data. Setworld_frame = odom_frameand execute local path plans and motions in this frame.Publishes
odom→rover_base_linktransform ifpublish_tf = true. -
GLOBAL EKF: Run another instance of a
robot_localizationstate estimation node that fuses all sources of data, including the GPS. Setworld_frame = map_frame. Use thenavsat_transform_nodeto transform the GPS data into an odometry message.Publishes
map→odomtransform ifpublish_tf = true.
The navsat transform node will provide the transform between odom and utm frames if datum parameter is provided.
Topics and nodes
First things first, the under the hood structure of the nodes:
Input data
As you can see, the raw data input is composed by the topics (the sensors are covered in next sections):
- /piksi_receiver/navsatfix_best_fix: a sensor_msgs/NavSatFix.msg that contains WGS 84 coordinates with best available solution at the moment (either RTK-FIX, RTK-FLOAT or SBAS). The data is in LLA (Latitude, Longitude and Altitude) in degrees and meters. Important values are the position_covariance and the status/status:
- If `position_covariance = 0.0049` and `status/status = 2`: it's RTK-FIX.
- If `position_covariance = 1` and `status/status = 2`: it's RTK-FLOAT.
- If `position_covariance = 25` and `status/status = 1`: it's SPP.
/piksi_attitude/baseline_heading: a piksi_rtk_msgs/BaselineHeading.msg that contains heading direction with best available solution at the moment. The data is inNEDframe and inmili-degrees (mdeg)./imu/data: a sensor_msgs/Imu.msg with the IMU raw data. The orientation is given inrad, the angular velocity inrad/sand the linear acceleration inm/s^2. Today (15/04/2020), the IMU package is wrong since it does not provide covariance values, hence they are hard-coded in a later step (explained in next sections).-
/tf: a tf/tfMessage.msg message containing all the required transformations for the localization, namely:- the robot model frames:
/scouting_base_link,/piksi_receiver,/piksi_attitude,/xsens. - the relation between the frames:
/mapand/odom(for now, this is constant).
- the robot model frames:
Output data
The result of this node is composed by the estimated position of the robot expressed in three different ways, the topics:
- /odometry/filtered/global: a nav_msgs/Odometry.msg message containing the filtered robot position in frame /odom->/scouting_base_link. This data is in cartesian coordinates.
/tf: a tf/tfMessage.msg message containing the/odometry/filtered/globaldata.gps/filtered: a sensor_msgs/NavSatFix.msg message containing the/odometry/filtered/globalin GPSWGS84coordinates.
Middle data
The extra topics result of processing the data along the way are the following:
- heading: a sensor_msgs/Imu.msg that contains heading direction with best available solution at the moment. The data is in ENU frame and in rad.
imu: a sensor_msgs/Imu.msg with the IMU modified data. The orientation is given inrad, the angular velocity inrad/sand the linear acceleration inm/s^2. The orientation is computed as the/headingdata updated with the variations of the orientation in the IMU. This transformation is performed because some experiments showed that the IMU magnetometer is not reliable inside the scouting box, hence, the long term readings weren't correct./odometry/gps: a nav_msgs/Odometry.msg message containing the GPS antenna position in frame/odom->/piksi_receiver. This data is incartesiancoordinates.
Nodes
The localization package is composed of 4 nodes (+1 which is not currently used):
- /heading_listener: this node takes in the topics /piksi_attitude/baseline_heading and /imu/data and process their data for the following nodes. It does the following processing:
1. Transforms the data in `/piksi_attitude/baseline_heading` from `NED` to `ENU` reference frame, and from `mdeg` to `rad`. Publishes it in the `/heading` topic.
2. Transforms the data in `/imu/data`, adding the covariance of the IMU for it's readings and modifying the `yaw` value: `yaw_imu = heading(last_available) + (imu_current - imu_0(when last_available heading))`. Publishes it in the `/imu` topic. The second modification may not be used depending on the configuration of the ekf.
/navsat_transform: this node transform the gps data provided by/piksi_receiver/navsatfix_best_fixto odometry data in/odometry/gps. It also converts back the filtered odometry in/odometry/filtered/globalto gps in/gps/filtered. The parameters used in this node are directly applied in the launch file:/launch/er_localization_node.launch./ekf_localization: this node computes the sensor fusion for the robot position. It requires data to be odometric and in ENU frame, hence, the conversions performed to the data in the previous nodes. The parameters of the ekf can be found in/cfg/ekf_imu_heading_localization.yaml/map_odom: for now, this node only provides a static transformation between the/mapframe and the/odomframe./set_datum: this node was used in a previous version to set thedatumparameter for the EKF node using a service server/client structure, however, in principle it is not required since the parameter is directly passed to it.
Sensors
For now (20/04/2020), the localization system is composed of 6 sensors: an IMU (Xsens MTi-3DK), 2 2.4Ghz radio (Freewave), and 3 RTK-GPS (Piksi Multi). One of the antennas, the Base Station, is fixed and used to send the corrections to the robot in order to have RTK quality. - Xsens MTi-3DK: is a high-precision IMU. Using the SDK, some information can be extracted but main configuration is done with the launch file of the driver node.
- Publishing frequency: 50 Hz
- USB-A 2.0 data transmission
-
Freewave radio: is a 2.4 GHz radio used to send/receive the Base Station data corrections. Currently, an Ethernet service is used in parallel for the same purpose.
- Serial cable RS232.
-
Piksi Multi: is a GPS antenna with RTK capability. The main configuration is done with the SDK of the sensor (see configuration of the sensor). The launch needs further investigation but, in general, it does not affect the internal configuration of the system.
- Publishing frequency
receiver: 10 Hz - Publishing frequency
attitude: 5 Hz - Ethernet communication.
- Serial cable RS232.
- Publishing frequency
File structure and content
The package file structure is as follows, two main directories:
1. /development_notes: has the documents published to ROS answers specifying some questions we had in the passed.
2. /earth_rover_localization: the ROS package.
/earth_rover_localization
-
/bag_files: currently not in use. Directory providing old data to do a simulation. -
/cfg: configuration files for the ROS nodes. -
/cmake: configuration file to find theGeographicLiblibrary, required for the/set_datumnode. -
/docs: image resources. -
/launch: launch files for the different available nodes. -
/mapviz_config: mapviz configuration file with latest debugging layout. -
/scripts: script to install thesbplibrary for the piksis. -
/src: contains the nodes' code.
/cfg
-
ekf_imu_heading_localization.yaml: configuration file for the robot localization EKF node. -
enu_origin.yaml: configuration for the Piksis initial position for the ENU conversion. -
mapviz_localization_config.yaml: deprecated configuration for mapviz. -
piksi_multi_driver_settings_rover_attitude.yaml: configuration file for piksi attitude node (pretty sure the configuration does not apply to the piksi). -
piksi_multi_driver_settings_rover_reference.yaml: configuration file for piksi reference node (pretty sure the configuration does not apply to the piksi). -
piksi_multi_driver_settings_roverserial.yaml: not used. -
xsens.yaml: configuration for the IMU.
/launch
-
er_localization_node.launch: to launch the localization along with the mapping system. -
er_localization_player.launch: Deprecated. Old launch of a simulated node (along with contents in/bag_files). -
er_localization_rover.launch: Deprecated. Old launch to use the localization node. -
er_localization_set_datum.launch: Deprecated. Old launch of the/set_datumnode. -
er_localization_viztools.launch: Deprecated. Old launch for mapviz. -
imu_piksi_drivers.launch: launches the sensors' drivers. -
mapviz.launch: launches mapviz. -
piksi_multi_rover_attitude.launch: launches the ROS driver node for the Piksi Attitude. -
piksi_multi_rover_reference.launch: launches the ROS driver node for the Piksi Receiver. -
xsens.yaml: launches the IMU ROS driver.
/mapviz_config
-
er_localization_config.mvc: configuration file for mapviz with the latest layout.
/scripts
-
install_sbp.sh: installs thesbplibrary.
/src
-
heading_listener.cpp: source code for the heading listener node. -
set_datum.cpp: Deprecated. Source code for theset_datumclient service.
Changelog for package earth_rover_localization
1.2.0 (2019-03-05)
- Update readme and install SBP script
- Fixed piksi package in launch and install targets
- Contributors: Andres Palomino
1.0.1 (2019-02-07) -----------* Update dependencies * Contributors: Andres Palomino
1.0.0 (2019-01-28) -----------* Update CMakeList.txt GeographicLib test * Update package.xml build depend * changing package name in launch file * adding dockerfile and docker-compose * Fix package name and update Readme * Contributors: Andres Palomino, Paul Harter
Wiki Tutorials
Source Tutorials
Package Dependencies
| Deps | Name | |
|---|---|---|
| 1 | rospy | |
| 1 | roscpp | |
| 1 | robot_localization | |
| 1 | tf | |
| 1 | piksi_multi_rtk | |
| 0 | earth_rover_xsens | |
| 2 | std_msgs | |
| 1 | piksi_rtk_msgs | |
| 1 | catkin | |
| 0 | earth_rover_common_messages | |
| 2 | nav_msgs | |
| 2 | sensor_msgs |
System Dependencies
Dependant Packages
Launch files
- launch/piksi_multi_rover_attitude.launch
-
- interface [default: tcp]
- baud_rate [default: 115200]
- tcp_addr [default: 192.168.8.223]
- base_station_ip [default: 192.168.8.1]
- load_enu_origin_from_file [default: false]
- launch/er_ekf_visual_odometry.launch
- Copyright (C) 2019 Earth Rover Limited. All rights reserved.
-
- color_image_raw [default: /camera0/color/image_raw]
- color_camera_info [default: /camera0/color/camera_info]
- depth_image [default: /camera0/depth/image_raw]
- depth_camera_info [default: /camera0/depth/camera_info]
- color_image_raw_rect [default: /image_rect_color]
- depth_image_rect [default: $(arg depth_image)]
- node_suffix [default: _0]
- bonding [default: --no-bond]
- launch/sbp_arbitrator.launch
- launch/rover_localization_node.launch
- Copyright (C) 2019 Earth Rover Limited. All rights reserved.
-
- launch/piksi_multi_rover_reference.launch
-
- interface [default: tcp]
- baud_rate [default: 115200]
- tcp_addr [default: 192.168.8.222]
- base_station_ip [default: 192.168.8.1]
- load_enu_origin_from_file [default: false]
- launch/visual_odometry_scouting.launch
-
- cam_suffix [default: _0]
- input_rgb_info [default: /camera$(arg cam_suffix)/color/camera_info]
- input_rgb_img [default: /camera$(arg cam_suffix)/color/image_raw]
- output_depth_rgb_img [default: /camera$(arg cam_suffix)/aligned_depth_to_color/image_raw]
- odom_topic [default: odom]
- launch/mapviz.launch
-
- print_profile_data [default: true]
- launch/autonomous_scouting_localization.launch
- Copyright (C) 2019 Earth Rover Limited. All rights reserved.
-
- launch/er_rtabmap_node.launch
- Copyright (C) 2019 Earth Rover Limited. All rights reserved.
-
- stereo [default: false]
- rtabmapviz [default: true]
- rviz [default: false]
- localization [default: false]
- use_sim_time [default: true]
- cfg [default: ]
- gui_cfg [default: ~/.ros/rtabmap_gui.ini]
- rviz_cfg [default: $(find rtabmap_ros)/launch/config/rgbd.rviz]
- frame_id [default: ugv_base_link]
- odom_frame_id [default: ]
- map_frame_id [default: ugv_base_link]
- ground_truth_frame_id [default: ]
- ground_truth_base_frame_id [default: ]
- namespace [default: rtabmap]
- database_path [default: ~/.ros/rtabmap.db]
- queue_size [default: 10]
- wait_for_transform [default: 0.2]
- args [default: ]
- rtabmap_args [default: $(arg args)]
- launch_prefix [default: ]
- output [default: screen]
- publish_tf_map [default: false]
- approx_sync [default: false]
- approx_sync [default: true]
- rgb_topic [default: /camera0/color/image_raw]
- depth_topic [default: /depth_registered/image_rect]
- camera_info_topic [default: /camera0/color/camera_info]
- depth_camera_info_topic [default: /depth_registered/camera_info]
- stereo_namespace [default: /stereo_camera]
- left_image_topic [default: $(arg stereo_namespace)/left/image_rect_color]
- right_image_topic [default: $(arg stereo_namespace)/right/image_rect]
- left_camera_info_topic [default: $(arg stereo_namespace)/left/camera_info]
- right_camera_info_topic [default: $(arg stereo_namespace)/right/camera_info]
- rgbd_sync [default: false]
- approx_rgbd_sync [default: true]
- subscribe_rgbd [default: $(arg rgbd_sync)]
- rgbd_topic [default: rgbd_image]
- depth_scale [default: 1.0]
- compressed [default: false]
- rgb_image_transport [default: compressed]
- depth_image_transport [default: compressedDepth]
- subscribe_scan [default: false]
- scan_topic [default: /scan]
- subscribe_scan_cloud [default: false]
- scan_cloud_topic [default: /scan_cloud]
- scan_normal_k [default: 0]
- visual_odometry [default: true]
- icp_odometry [default: false]
- odom_topic [default: odom]
- vo_frame_id [default: $(arg odom_topic)]
- publish_tf_odom [default: true]
- odom_tf_angular_variance [default: 1]
- odom_tf_linear_variance [default: 1]
- odom_args [default: ]
- odom_sensor_sync [default: false]
- odom_guess_frame_id [default: ]
- odom_guess_min_translation [default: 0]
- odom_guess_min_rotation [default: 0]
- imu_topic [default: /imu/data]
- wait_imu_to_init [default: false]
- subscribe_user_data [default: false]
- user_data_topic [default: /user_data]
- user_data_async_topic [default: /user_data_async]
- gps_topic [default: /gps/fix]
- tag_topic [default: /tag_detections]
- tag_linear_variance [default: 0.0001]
- tag_angular_variance [default: 9999]
- rgb_topic_relay [default: $(arg rgb_topic)_relay]
- rgb_topic_relay [default: $(arg rgb_topic)]
- depth_topic_relay [default: $(arg depth_topic)_relay]
- depth_topic_relay [default: $(arg depth_topic)]
- left_image_topic_relay [default: $(arg left_image_topic)_relay]
- left_image_topic_relay [default: $(arg left_image_topic)]
- right_image_topic_relay [default: $(arg right_image_topic)_relay]
- right_image_topic_relay [default: $(arg right_image_topic)]
- rgbd_topic_relay [default: $(arg rgbd_topic)]
- rgbd_topic_relay [default: $(arg rgbd_topic)_relay]
- launch/er_localization_node.launch
- Copyright (C) 2019 Earth Rover Limited. All rights reserved.
-
- mounting [default: tractor]
- launch/sbp_logger.launch
- launch/scouting_localization_node.launch
- Copyright (C) 2019 Earth Rover Limited. All rights reserved.
-
- launch/dual_xsens.launch
-
- cam_suffix [default: _0]
- launch/xsens.launch