I spent all day working on getting the gps fusion problems worked out of Jackal. I think I made major progress.

1. I found that the configuration of nodes from navsat_tranform to ekf was not correct.

   1. jackals ekf should not have been used at the filtered odometry input for navsat it was changed to the output of the ekf form the gps. See links to source code and explinations.
      https://github.com/cra-ros-pkg/robot_localization/blob/kinetic-devel/launch/dual_ekf_navsat_example.launch
      https://github.com/cra-ros-pkg/robot_localization/blob/kinetic-devel/params/dual_ekf_navsat_example.yaml
      http://docs.ros.org/kinetic/api/robot_localization/html/integrating_gps.html

2. I have also found what might be the link to allow me to have the move base system move to gps goals

   1. when using the navsat transform i have it set to publish a UTM transform relative to the map. This means if I give the system a utm point it should be able to show it on the map frame and relative to anything else.

   2. In this link below it is stated that I need to figure out what frame to set the navigation system to. Far more research into the navigation stack (move base in particular) will be need.
      http://answers.ros.org/question/202117/integrate-a-gps-sensor-with-robot_localization-and-use-move_base-node/

3. I have also been able to write code that can take lat, long, height and convert it into UTM pose for ROS

   1. this will be needed if the user wants to send goals to the robot

   2. the output of this should be piped into move_base once that is working

4. In conjunction with the code to UTM transform I found how to send a goal to the planning system. It is found on page 362 of "Learning ROS for robotics programming"

5. Need to figure out what is causing this orientation problem, I have a feeling that what I did to the gps fusion system should resolve it, or I hope so.