A selection of best-performing designs generated with RoboGrammar for four different terrains.
CMake >= 3.8
- Ubuntu: The version available through
apt-getis probably outdated, install from the link above
GLEW
- Ubuntu:
sudo apt-get install libglew-dev
Python 3.6 + headers
If you are on an older version of Ubuntu, you may need to add the "deadsnakes" PPA first:
sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt-get update
- Ubuntu:
sudo apt-get install python3.6 python3.6-dev
git clone git@github.com:allanzhao/robot_design.git
cd robot_design
git submodule update --init
mkdir build; cd build
cmake -DCMAKE_BUILD_TYPE=RelWithDebInfo -DPYTHON_EXECUTABLE=/path/to/python3.6 .. (replace /path/to/python3.6 as appropriate)
make -j4 (replace 4 with the number of CPU cores available)
Using a virtualenv is recommended.
Add the robot_design/build/examples/python_bindings and the robot_design/examples/design_search directories to your PYTHONPATH environment variable, and make sure you are in the robot_design directory.
virtualenv -p python3.6 venv
source venv/bin/activate
pip3 install numpy numpy-quaternion
Make sure you are in the build directory created earlier.
View an example robot design:
examples/viewer/Viewer ../data/designs/insect.dot 0 1 1 1 2 2 2 3 3 3 5 5 5 4 4 4 -r
Optimize and view a trajectory for the example robot:
examples/viewer/Viewer ../data/designs/insect.dot 0 1 1 1 2 2 2 3 3 3 5 5 5 4 4 4 -o -r
View a rule from the grammar:
examples/rule_viewer/RuleViewer ../data/designs/insect.dot 0 rhs -r (views the right-hand side of rule 0)
Make sure you are in the robot_design directory and the virtualenv is active (source venv/bin/activate).
Optimize and view a trajectory for the example robot:
python3 examples/design_search/viewer.py FlatTerrainTask data/designs/grammar_jan21.dot -j16 0, 6, 20, 12, 2, 7, 18, 20, 10, 4, 20, 10, 11, 5, 10, 4, 10, 5, 19, 5 -o