Inverse Kinematics
Overview
Inverse kinematics (IK) finds joint configurations that place a robot’s end effector, or an articulated body’s limbs, at desired poses; it is central to manipulation, teleoperation, motion capture, and animation. Classical approaches include geometric and algebraic solutions, Jacobian-based methods such as damped least squares, and constrained optimization with null-space control for redundancy, joint limits, and collision avoidance. Learned methods use neural networks, graph models, and generative models to produce fast or multiple candidate solutions, often combined with classical refinement. A key challenge is balancing speed and accuracy with smooth motion and safety near singularities, while making learned solvers generalize across robots and difficult or unreachable targets.
Latest papers
40 papers4 papers in the last four weeks, down 43% on the four weeks before. 0.0% of all new papers.