The core python library of the Giskard framework for constraint- and optimization-based robot motion control.
Do this:
sudo pip3 install pybullet
sudo pip3 install scipy
sudo pip3 install casadi
sudo pip3 install sortedcontainers
sudo pip3 install hypothesis
sudo pip3 install pandas
sudo pip3 install numpy
sudo apt install python3-dev
Install one of the following QP solver. The solvers are ordered by how fast they can solve the problem constructed by Giskard. QPOases is the fastest opensource solver for my usecase, that I have found. However, it is still significantly slower than the other two options:
- Gurobi:
sudo pip3 install gurobipy- If you have vpn access or are in the local network of the IAI of the University of Bremen, follow these instructions: https://ai.uni-bremen.de/wiki/intern/adm/gurobi
- Otherwise you can apply for a free academic license or buy a license here: https://www.gurobi.com/academia/academic-program-and-licenses/
- cplex
- Check this PR for more information: #77
- QPOases
- Install noetic branch of https://github.com/SemRoCo/qpOASES
Finally, you will need to set the qp_solver name in the config file, you are loading. E.g. here: https://github.com/SemRoCo/giskardpy/blob/mpc/config/default.yaml.
Now create the workspace
source /opt/ros/<ros-version>/setup.bash # source ROS
mkdir -p ~/giskardpy_ws/src # create directory for workspace
cd ~/giskardpy_ws # go to workspace directory
catkin init # init workspace, you might have to pip install catkin-tools
cd src # go to source directory of workspace
wstool init # init rosinstall
wstool merge https://raw.githubusercontent.com/SemRoCo/giskardpy/devel/rosinstall/<ros-version>.rosinstall
# update rosinstall file
wstool update # pull source repositories
rosdep install --ignore-src --from-paths . # install dependencies available through apt
cd .. # go to workspace directory
catkin build # build packages
source ~/giskardpy_ws/devel/setup.bash # source new overlay
Giskard will run much faster with Adrian Röfers bullet bindings instead of the official ones.
./scripts/build_pybullet.sh /my/awesome/library/dir
If everything worked fine, you should be able to do:
import betterpybullet as bpbAfterwards you can set the collision checker to bpb in config/default.yaml.
collision_avoidance:
collision_checker: bpb # one of [bpb, pybullet, disable]