Pinker¶
Python inverse kinematics for embedded robotics.
Inverse kinematics in Pinker is defined by weighted tasks,
limits and barriers. Once a robot is loaded,
its current geometric state is held in its configuration. Given a configuration, tasks and a time step,
pinker.solve_ik.solve_ik() computes joint velocities that steer the model
towards fulfilling all tasks at best:
velocity = solve_ik(configuration, tasks, dt, solver="daqp")
configuration.integrate_inplace(velocity, dt)
To get started, install Pinker and a QP solver of your choice, then follow the first script in Examples. The Introduction sets out notations and introduces the two main concepts of robot configuration and task used in the library.
Getting started
API documentation