You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently the ros_reasoner doesn't take a node name into account for changing the system_mode of a node. This is supposed to happen in the request_configuration function, but this is not being done. One way to fix this for now is to maybe query the ontology for which components solve a fd, or these nodes need to be mapped directly to the fds. In the first case the node names would need to match the names of the components. As this is a bit of a conceptual decision I think it is good to maybe discuss this.
The text was updated successfully, but these errors were encountered:
Actually, this concept of node name is related to system modes, so I would suggest changing the '/ros_reasoner/change_node_mode' service to request a function design for a specific function.
So, I suggest changing the service msg NodeMode to something like this:
Then in the system modes bridge we can have a logic to select the node to be changed. My suggestion is to use the function name as the default node name. Then, if we want to use a custom mapping, we could have a dict that maps functions and fds to different node names.
Currently the ros_reasoner doesn't take a node name into account for changing the system_mode of a node. This is supposed to happen in the request_configuration function, but this is not being done. One way to fix this for now is to maybe query the ontology for which components solve a fd, or these nodes need to be mapped directly to the fds. In the first case the node names would need to match the names of the components. As this is a bit of a conceptual decision I think it is good to maybe discuss this.
The text was updated successfully, but these errors were encountered: