Python API

cuRobo provides accelerated modules for robotics which can be used to build high-performance robotics applications. The library has several modules for numerical optimization, robot kinematics, geometry processing, collision checking, graph search planning. cuRobo provides high-level APIs for performing tasks like collision-free inverse kinematics, model predictive control, and motion planning.

High-level APIs:

• Motion Generation / Planning: curobo.wrap.reacher.motion_gen.

• Inverse Kinematics: curobo.wrap.reacher.ik_solver.

• Model Predictive Control: curobo.wrap.reacher.mpc.

• Trajectory Optimization: curobo.wrap.reacher.trajopt.

cuRobo package is split into several modules:

• curobo.opt contains optimization solvers.

• curobo.cuda_robot_model contains robot kinematics.

• curobo.curobolib contains the cuda kernels and python bindings for them.

• curobo.geom contains geometry processing, collision checking and frame transforms.

• curobo.graph contains geometric planning with graph search methods.

• curobo.rollout contains methods that map actions to costs. This class wraps instances of curobo.cuda_robot_model and curobo.geom to compute costs given trajectory of actions.

• curobo.util contains utility methods.

• curobo.wrap adds the user-level api for task programming. Includes implementation of collision-free reacher and batched robot world collision checking.

• curobo.types contains custom dataclasses for common data types in robotics, including JointState, CameraObservation, Pose.