This library implements differentiable robot tree from URDF or MCJF robot format, and the differentiable planning objects such as obstacle avoidance, self-collision avoidance and via point.
Simply do
pip install -e .For benchmarking on computation time of all available robot kinematics
python examples/forward_kinematics.py
For benchmarking on computation time of distance fields
python examples/collision_distance.py
A part of this implementation is inspired from the library differentiable robot model.
If you have any questions or find any bugs, please let me know: An Le, an@robot-learning.de
If you are using TorchRobotics for your scientific publications, please cite it using the CITATION file, and the github help page.