The site is read-only. Please transition to use Robotics Stack Exchange
| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
 | 1 | initial version |
What you describe sounds like behaviour modelling, which can be done with either an hierarchical state machine (HSM), or a behaviour tree. While it can be done by starting and stopping different nodes, I'd say that a HSM which coordinates the activation of nodes would be much simpler.
| | 2 | No.2 Revision |
What you describe sounds like behaviour modelling, which can be done with either an hierarchical state machine (HSM), (HSM) for instance, or a behaviour tree. While it can be done by starting and stopping different nodes, I'd say that a HSM which coordinates the activation of nodes would be much simpler.
| | 3 | No.3 Revision |
What you describe sounds like behaviour modelling, which can be done with an hierarchical state machine (HSM) for instance, or a behaviour tree. While it can be done by starting and stopping different nodes, I'd say that a HSM which coordinates the activation of nodes would be much simpler.
Edit:
Thank you @gvdhoorn i just googled HSM + ROS and came up with SMACH. just to be clear is that the kind of approach that you are suggesting for this job?
SMACH is indeed the go-to HSM implementation in ROS. But there are probably many more.
For behaviour trees specifically, see for instance github.com/futureneer/beetree (unfinished).
For all sorts of behaviour modeling tools, see wiki/decision_making.
These are just examples, I don't necessarily particularly recommend any of these.
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
