timestamp on the message is earlier than all the data in the transform cache'

[Pito Salas]

Hi friends,

I am getting this message… Somewhere in the back of my memory I thought something like this (or related to this) has come up on this list but I could not locate it. I am currently getting it from (or via) slam_toolbox but I don’t feel it is because of slam_toolbox:

Message Filter dropping message: frame 'laser' at time 1763085485.168 for reason 'the timestamp on the message is earlier than all the data in the transform cache'

I'm trying to sort out this message, and whether I can ignore it. From my study and troubleshooting, the laser tf is static, built from this in my urdf:

```

  <joint name="base_link_to_laser" type="fixed">

    <parent link="base_link"/>

    <child link="laser"/>

    <origin xyz="${laser_x_offset} 0 ${laser_z_offset}" rpy="0 0 ${-pi/2}"/>

  </joint>

```

Because it is static, the time stamp of that tf will refer back to when I first started the 

```

        Node(

            package='robot_state_publisher',

            executable='robot_state_publisher',

            name='robot_state_publisher',

            output='screen',

            parameters=[

                {

                    'use_sim_time': LaunchConfiguration('use_sim_time'),

                    'robot_description': Command(['xacro ', LaunchConfiguration('urdf')])

                }

            ]

        ),

```

So it is "normal" that the timestamp is "earlier than all the data in the transform cache". And I can ignore it. 

Am I thinking about this the right way? 

Best,

Pito

Boston Robot Hackers && 

Comp. Sci Faculty, Brandeis University (Emeritus)

[James H Phelan]

Pito,

I had this same problem.  Google was no help.  AI was no help.

Mike Wimble and I jumped through all kinds of hoops to figure this out and it eventually resolved.

But I can't remember what exactly fixed it.  I think it was something about conflicting publishers.

Or Stamped vs Unstamped cmd_vel.

I'm going to cc this to Mike and see if he remembers where the gremlin was.

Sounds like you're just behind me climbing Mt. Learning Curve!

Just last weekend I was finally able to map & navigate the house successfully.

For your amusement:

https://github.com/JHPHELAN/stingray

Overview of Stormy the Stingray.

Scroll to the very bottom of README for YouTube links.

It's best to open them in separate side-by-side windows.  

Run the Rviz screencast, then when you see the green goal arrow, start the iPhone video so they track together.

"When the student is ready, a teacher appears."

Your recent appeal for best practices spurred a great response.  Thank you!

Having finally achieved a bumbling level of success, I too am ready to ask "how is this supposed to be done?"

Sergei is superb at defining the (or at least a) "right way" to do things.

His AI request to summarize proven code was brilliant and a great resource. 

(Now I just have to get an AI to explain it to me!)

Mike Wimble is an excellent teacher and I am going to start reading his book "ROS for Mere Mortals":

https://github.com/wimblerobotics/Robotics_Book/blob/main/book/chapters/book.pdf

James

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

--
You received this message because you are subscribed to the Google Groups "HomeBrew Robotics Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hbrobotics+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/hbrobotics/964DA9BB-E698-4EC6-9ED9-1533DB5D717E%40gmail.com.