This project is based on a Kinova robotic arm URDF converted to an MJCF format for MuJoCo simulation. It integrates Pinocchio for dynamic computations and develops demos for kinematics, trajectory interpolation, and control. The goal is to create a MuJoCo-based, cross-platform equivalent of MoveIt! for comprehensive motion planning and control.
It is recommended to use a conda environment. Here is an example of creating a new environment.
conda create --name Mujoco-Kinova python=3.10
conda activate Mujoco-Kinova
python.exe -m pip install --upgrade pip setuptools wheelThe corresponding dependencies need to be installed below; if all goes well, these two lines of commands should be sufficient.
# conda install pinocchio
conda install pinocchio -c conda-forge
# python requirements
pip install mujoco dearpygui scipy numpy pandas matplotlib scikit-learn PyQt6 -i https://pypi.tuna.tsinghua.edu.cn/simpleIn file ./Kinematics/Inverse_Demo/1_5_inverse_impedance_control.py, achieved together control of least_square based Kinematics and Pinocchio Dynamics
Further research will be conducted on 7-dof issue of multiple analytical solutions.
