FootstepStatusMessage isn't populated correctly.

[Maurice Fallon]

• the sequence of messages below published: one with pos/orientations are filled, then immediately one with zeros.
• robot_side is never updated - its always 0. This is certainly wrong.
• unique_id is never updated. (not important)

status: 0 footstep_index: 0 robot_side: 0 actual_foot_position_in_world: x: 0.0 y: 0.0 z: 0.0 actual_foot_orientation_in_world: x: 0.0 y: 0.0 z: 0.0 w: 0.0 is_done_walking: False unique_id: 0

---

status: 1 footstep_index: 0 robot_side: 0 actual_foot_position_in_world: x: 0.396249815921 y: 0.0982950787699 z: 0.244278684006 actual_foot_orientation_in_world: x: -0.00902732712482 y: -0.00661898549606 z: -0.005820477068 w: 0.999920406054 is_done_walking: False unique_id: 0

---

status: 0 footstep_index: 1 robot_side: 0 actual_foot_position_in_world: x: 0.0 y: 0.0 z: 0.0 actual_foot_orientation_in_world: x: 0.0 y: 0.0 z: 0.0 w: 0.0 is_done_walking: False unique_id: 0

---

status: 1 footstep_index: 1 robot_side: 0 actual_foot_position_in_world: x: 0.48352975188 y: -0.105733820017 z: 0.244477975626 actual_foot_orientation_in_world: x: 0.0101170436445 y: -0.00564744280715 z: -0.0674450410532 w: 0.997655711283 is_done_walking: False unique_id: 0

---

[Maurice Fallon]

FYI @simalpha

[DouglasS]

@simalpha @mauricefallon For which robot(s) do you see this behavior?

[Maurice Fallon]

For both AtlasFinalsROSAPISimulator and ValkyrieROSAPISimulator

[DouglasS]

Is there any useful output in the ROS logs or the IHMC error ROS topic?

[Maurice Fallon]

You can see the issue by monitoring the data published when walking with:

rostopic echo /ihmc_ros/valkyrie/output/footstep_status

Basically you see these empty messages being regularly published and that the robot_side bit is never set to be right.

In the link below you can find a rosbag which will send some footsteps commands so you can test it on your end:

1. Launch ValkyrieROSAPISimulator

2. Play back the ROS log to make the robot walk 6 steps: rosbag play footstep_status.bag

3. use rostopic echo to monitor FootstepStatusMessage

https://www.dropbox.com/sh/9724mrpdpbbh7wi/AABYYzulQWJwGK_SqNxo8V_Da?dl=0

Also attached is a spreadsheet with the errors pointed out.

[DouglasS]

@nmertins Did you encounter this while you were doing tests for the new Atlas release last week?

@mauricefallon thanks for the data, I was more curious as to the content of the error topics though as that is the best way to identify possible problems like mismatches between resources in the catkin workspace and resources in the distribution binaries, or as the case may be the source code/IDE workspace. We'll take a look at this more seriously over the next few days and see if we can track it down.

[Nathan Mertins]

I didn't see any problems running the real robot. I'll double check this ASAP.