Simply run roslaunch ada_launch default.launch.
Next, you can run a demo from ada_demos, that uses the plain Ada robot without any attachments.
The feeding demo requires some more configuration and some additional nodes.
It can be started by running roslaunch ada_launch default.launch feeding:=true.
To run it with simulated perception roslaunch ada_launch default.launch feeding:=true perception:=false.
You may also need to start
- The camera node, running directly on the camera board. Typically the script is called something like
run_all.sh - The forque sensor controller, using
roslaunch forque_sensor_hardware forque.launch - The feeding demo parameters, and the feeding demo itself.
Simply run roslaunch ada_launch simulation.launch.
This adds mandatory TF trees that the ADA demos are expecting, plus simulated food and face detection.
Run roslaunch ada_launch simple_perception.launch.
It takes one argument adareal, which should be set to true if running on the real robot:
roslaunch ada_launch simple_perception.launch adareal:=true
For more information, see the simple_perception demo in ada_demos.