Posted by @tvrzjak:
Hello,
First of all I would like to thank you for the great work. But I have a small problem. I was able to run the simulation demo examples, I run my own simulation, but I’m not exactly sure how to connect to my real robot. I use the DJI RoboMaster EP with this slightly modified ROS2 driver: GitHub - jeguzzi/robomaster_ros: ROS2 for DJI Robomaster EP and S1. What it actually does is that it connects to the DJI control unit and its python API and through it publish the read information (e.g. encoders) to ROS and, on the other hand, writes subscribed messages (e.g. wheel speeds) to the robot through the API. I also use nav2 library and slam toolbox.
I know I need to write my own adapter, but I would actually just need to connect to the already finished DJI ROS2 driver. Alternatively, just subscribe and publish classic topics in ROS navigation (base_link, cmd_vel, etc.). Is it possible to connect rmf to the robot even just in this way, and if so, how?
Posted by @mxgrey:
There are two potential answers to your question depending on whether your robot’s setup is using nav2, the standard navigation stack for ROS 2.
If yes, then I would recommend using free_fleet to connect your robot to Open-RMF.
If not, then the usual method is to use rmf_fleet_adapter_python to write a Python script that connects your robot’s API to RMF. You can refer to rmf_demos_fleet_adapter for a working example of what the source code might look like. That’s the fleet adapter that gets used in the demo simulations. If you choose to write the fleet adapter open source, we may be able to help guide and review its development.
Posted by @tvrzjak:
Thank you. I’m using the nav2 navigation stack, so it should be enough to just fill in the Client_node_config.hpp and Server_node_config.hpp and communication with the navigation stack is already ensured?
Posted by @Blvckb0x:
Hi sir, i hope u r doing well, im working on the same subject, is it possible to share ur code or github link that i can use to apply the slam technique to my EP. Thank you