ROS-LEAFLET-GPS

This application is designed to display an OSM map with Leaflet in a browser, while connected to ROS, so it can display the position and trajectory of the GPS (/NavSatFix) using a marker.

This package is also able to integrate visual odometry results in geometry_msgs/PoseStamped format.

On the left side you can see the ruler like button. You can use it to measure the metric distance.

A screenshot of the demo program. The map marker represents GPS results and the vehicle represents the estimated pose using visual odometry.

Thanks to:

1. ROS-OSM-map-integration for providing the ROS interface.
2. Leaflet PolyLine Measurement Plugin for providing the measurement plugin.

Installation

This is not actually a ROS package, so you do not need to put it in catkin workspace. It connects to ROS to get information from GPS sensor, and uses some ROS functionality, for instance ROS param, ROS topic, etc.

To allow this application to connect with ROS, you need to install RosBridge, which is kind of a bridge to connect some Javascript apps to the ROS core. To do so, please follow the tutorial.

Simply type:

 sudo apt install ROS-<ROS-VERSION>-rosbridge-suite

In order to launch rosbridge with the rest of your stuff, you can add this to your launch file :

    <launch>
        <include file="$(find rosbridge_server)/launch/rosbridge_websocket.launch" > 
            <arg name="port" value="9090"/>
        </include>
    </launch>

And then, open the page index.html with your favorite browser, and you're ready. You can even set the default page of your browser to it, so it opens automatically on startup.

Structure

    .
    ├── assets                  # Pictures and icons.
    ├── launch                  # Demo launch files
    ├── lib                     # All important js libraries, including leaflet, leaflet plugins, three 3D lib.
    ├── index.html              # Demo html page for displaying the map.
    ├── script_pose.js          # Demo js script to display `geometry_msgs/PoseStampedWithCovariance` message.
    ├── script_fix.js           # Demo js script to display `NavSatFix` for GPS 
                                # and `rds_msgs/msg_novatel_bestpos` for self defined pose.
    ├── script_reuse_path.js    # Demo js script to display `PoseStampedWithCovariance` message
                                # for GPS and `nav_msgs/Path` for self defined pose.
    └── README.md

Demo

You can download the ROS bag to directly try the demo. In order to make the two trajectories generated by SLAM and GPS discriminative, I intented to make the SLAM not so accurate.

This ROS bag contains GPS pose (/gps_pose), SLAM pose (/sslam_estimator_node/camera_pose) and SLAM path (/pose_graph_path). For the full bag containing image information, please download the link here.

Configuration

The configuration of the module can be modified just by changing the value at the top of the script.js file.

• CONFIG_tile_source : Set the source of the tiles for the map. If you downloaded the maps of the area you want to move in, then you can set it to local. Else, set it to server.
• CONFIG_tile_local_path : Path to the downloaded tiles
• CONFIG_ROS_server_URI : Route to ROS server. It could be localhost or an IP.

Parameters

In the lab, we're working with two different GPS, which are not publishing on the same topic and at the same frequency. So two parameters can be set :

• /panel/gps_topic for the GPS topic's name. Example : /gps
• /panel/pose_topic for the VO topic's name. Example : /slam/pose
• /panel/nb_cycles for the number of cycles between each refreshing. Example : 20

You can set those in a ROS launch file, or set it using rosparam set, then refreshing the page in the browser by clicking on the refresh button.

Warning!

Updates on 2020-12-15

You need first to know your city and search the corresponding UTM coordindates in https://www.gps-latitude-longitude.com/. You can find the keyword "zone" like below:

Just replace the '48' and 'N' in script_reuse_path.js utm functions with the zone (56) and band (H).

Example 1

1. Run roslaunch ./launch/novatel_pose.launch

2. Open index_pose.html with a browser

3. Play demo ROS bag file

This will display GPS (geometry_msgs/PoseStampedWithCovariance) in red trajectory and estimated pose (geometry_msgs/PoseStampedWithCovariance) in blue trajectory.

Example 2

1. Run roslaunch ./launch/novatel_path.launch

2. Open index_path.html with a browser

3. Play demo ROS bag file

This will display GPS (geometry_msgs/PoseStampedWithCovariance) in red trajectory and estimated pose (nav_msgs/Path) in blue trajectory.

The advantage of nav_msgs/Path over geometry_msgs/PoseStampedWithCovariance is that it is not fixed. Pose graph optimization and loop closure might update your SLAM trajectory. The novatel_path is able to update the drawn path of your estimated pose.