This project requires ROS to run, which should already be installed on several of the clic machines.
In addition to ROS, you will also need to install Portaudio, Pyaudio, and SpeechRecognition. Here is how to do it to your user directory on a clic machine:
- Download portaudio from http://www.portaudio.com/download.html and save to folder portaudio in your home directory
- ./configure --prefix=/home/UNI/portaudio_path make make install
- Open ~/.pydistutils.cfg and save the following in that file: [build_ext] include_dirs=/home/UNI/portaudio_path/include/ library_dirs=/home/UNI/portaudio_path/lib/
- pip install --user --allow-external pyaudio --allow-unverified pyaudio pyaudio
- Pip install --user SpeechRecognition
First, clone this repo.
For every new terminal you open, you will need to run:
$ source load_settings.shTo set the path in the kinect processor file, run
$ python create_paths.pyfrom the project directory.
To generate the rest of the block models from block A, go into the models folder [baxter-abcs/src/system_launch/world/models] and run:
$ python generate_block_models.pyYou should only need to do this once.
Make sure you catkin_make before running the project.
In separate terminals...
$ source load_settings.sh
$ roslaunch system_launch everything.launch$ source load_settings.sh
$ rosrun baxter_tools enable_robot.py -e
$ rosrun vision kinect_processor$ source load_settings.sh
$ rosrun control control.py$ source load_settings.sh
$ rosrun control pick_up_blocks.py$ source load_settings.sh
$ rosrun speech_command speech_command.py$ killall gzserver when system_launch dies and refuses to restart
$ ps aux | grep 'ros' if the error 'port in use' comes up, run this to make sure no leftover ros processes, if there are any 'kill_all_ros.sh' should solve the problem
$ rosrun hello_baxter hello_baxter.py run a script within a custom package