Search Results

Also, if you use the utm_odometry_node in the gps_common package, you'll get UTM coordinates for each of your GPS positions. … Offhand, I don't know of any existing ROS nodes that extract that information from GPS units. You might want to go the route suggested by @ahendrix. …

Quick question: Am I correct in assuming the rtk-fixed GPS messages go as nav_msgs/odometry into a robot_localization_node after passing through a navsat_transform_node? Thanks! …

Did you compile this from source or use the package manager? When Groovy was EOL'd, I stopped updating its source and releases. Having said that, if you pull down the Hydro or Indigo versions, it oug …

Third - and this is critical - if you need to fuse the GPS data using ekf_localization_node, do not use the set_pose topic/service for feeding in the GPS data to ekf_localization_node. … , and as it is shown in my ekf_template.launch file I haven't included the GPS. …

Note, however, that some GPS devices provide heading based on differentiated positions, and their ROS drivers would likely fill out those fields in the messages they produce. …

While I can generate a made-up bag file with the right GPS coordinates to test this situation, if you happen to have one that you can share, I'd like to use it. …

But I still believe that it is possible to compute the absolute velocity(what I need) by absolute positions(GPS). is it right? …

The state estimation nodes in r_l don't care about GPS data; the tutorials just use GPS data as a means of showing how to work with two state estimation node instances (one each for the map and odom frames … This node is not required for operation of the state estimation nodes, and only serves to allow users to work with GPS devices. …

If you lose GPS signal, then yes, you will continue to get a state estimate in your map instance of the EKF using just the odometry and IMU data. …

In the meantime, what was your first GPS message? If you set a datum, navsat_transform_node assumes that datum is your robot's "initial" GPS reading. … So if your first GPS reading was tens of meters away from 31.2630568333, 121.616275333, then the output will reflect that. …

Integration of GPS with robot_pose_ekf proved to be difficult for me when I tried it as well. … The former can be used in place of robot_pose_ekf, and the latter allows for integration of GPS data. The tutorials page contains a tutorial that details how to integrate a GPS sensor. …

For @jcerruti, I'm wondering if the issue is the third parameter, which is pulled directly from the message's header (in your case, the GPS data message header). … Could your GPS driver be pulling that directly from the GPS unit, and not from your machine? Just a thought. @mikepurvis, what sensor data message is causing this? …

What are the covariances of the state estimate and GPS measurements? … I'll be adding new parameters to handle GPS data and still comply with REP-105. …

Depending on your GPS, there will likely be some extra effort required to get its data integrated. Those nodes will generate a odom->base_link transform for you. … However, if you really want to follow the ROS spec (particularly REP-105), you'll need to maintain all three frames (map, odom, and base_link), and only fuse GPS data with the estimate in the map frame …

I'm very suspicious of your GPS data. … Can you post the value of /odometry/gps? …