Hydro move_base/tf2 extrapolation error into the future errror
Hi All,
We are using Hyrdo's move_base and we continue to get an extrapolation error into the future. I believe this is causing our robot to stop and start during a traverse.
In looking the ROS source code it appears that the lookupTransform in navigation/base_local_planner/src/goal_functions.cpp line 110 in transformGlobalPlan is returning the error. We are using the dwa_local_planner. We are publishing transforms at 20hz from a separate node. The parameters to tf.lookupTransform are global_frame = "local" and the plan_pose.header.frame_id = "global", and the plan_pose.header.stamp comes from the generating the gloabl plan. I think the bug is coming from tf2/src/buffer_core.cpp in WalkToTopParent(). I still need to do more debugging to find the exact problem. I've turned the debug information on for move_base so I'm including a snippet of it, and the console log below. Any help would be appreciated. Thanks.
[DEBUG] [1386166204.375994679]: The action server has received a new goal request
[DEBUG] [1386166204.376260355]: A new goal has been recieved by the single goal action server
[DEBUG] [1386166204.376656845]: Accepting a new goal
[DEBUG] [1386166204.376895493]: Accepting goal, id: /waypoint_management_node-1-1386166204.372833387, stamp: 1386166204.37
[DEBUG] [1386166204.377491239]: Publishing feedback for goal, id: /waypoint_management_node-1-1386166204.372833387, stamp: 1386166204.37
[DEBUG] [1386166204.377705512]: Publishing feedback for goal with id: /waypoint_management_node-1-1386166204.372833387 and stamp: 1386166204.37
[DEBUG] [1386166204.378025665]: Planning...
[DEBUG] [1386166204.378168840]: Waiting for plan, in the planning state.
[DEBUG] [1386166204.378332478]: Full control cycle time: 0.000979315
[DEBUG] [1386166204.455307564]: Got Plan with 49 points!
[DEBUG] [1386166204.457263957]: Generated a plan from the base_global_planner
[DEBUG] [1386166204.458307175]: Planner thread is suspending
[DEBUG] [1386166204.477544259]: Publishing feedback for goal, id: /waypoint_management_node-1-1386166204.372833387, stamp: 1386166204.37
[DEBUG] [1386166204.477708316]: Publishing feedback for goal with id: /waypoint_management_node-1-1386166204.372833387 and stamp: 1386166204.37
[DEBUG] [1386166204.477914069]: Got a new plan...swap pointers
[DEBUG] [1386166204.478065066]: pointers swapped!
[ INFO] [1386166204.478229751]: Got new plan
[DEBUG] [1386166204.478925371]: In controlling state.
[DEBUG] [1386166204.515009849]: Got a valid command from the local planner: 0.500, 0.000, 2.000
[DEBUG] [1386166204.516475469]: Full control cycle time: 0.037907056
[DEBUG] [1386166204.578542367]: Publishing feedback for goal, id: /waypoint_management_node-1-1386166204.372833387, stamp: 1386166204.37
[DEBUG] [1386166204.579298889]: Publishing feedback for goal with id: /waypoint_management_node-1-1386166204.372833387 and stamp: 1386166204.37
[DEBUG] [1386166204.580845315]: In controlling state.
[DEBUG] [1386166204.609181880]: Got a valid command from the local planner: 0.500, 0.000, 2.000
[DEBUG] [1386166204.610400540]: Full control cycle time: 0.031062960
[DEBUG] [1386166204.677557610]: Publishing feedback for goal, id: /waypoint_management_node-1-1386166204.372833387, stamp: 1386166204.37
[DEBUG] [1386166204.677725508]: Publishing feedback for goal with id: /waypoint_management_node-1-1386166204.372833387 and stamp: 1386166204.37
[DEBUG] [1386166204.677947673]: In controlling state.
[DEBUG] [1386166204.704567051]: Got a valid command from the local planner: 0.500, 0.000, 2.000
[DEBUG] [1386166204.705445516]: Full control cycle time: 0.027409982
[DEBUG] [1386166204.777927638]: Publishing feedback for goal, id: /waypoint_management_node-1-1386166204.372833387, stamp: 1386166204.37
[DEBUG] [1386166204.779469173]: Publishing feedback for goal with id: /waypoint_management_node-1-1386166204.372833387 and stamp: 1386166204.37 ...