NODEIK: Solving Inverse Kinematics with Neural Ordinary Differential Equations for Path Planning

This paper proposes a novel inverse kinematics (IK) solver of articulated robotic systems for path planning. IK is a traditional but essential problem for robot manipulation. Recently, data-driven methods have been proposed to quickly solve IK for path planning. These machine learning-based models can handle a large amount of IK requests at once by leveraging the GPU. However, such methods suffer from reduced accuracy and considerable training time. We propose an IK solver that improves accuracy and memory efficiency with continuous normalizing flows by utilizing the continuous hidden dynamics of a Neural ODE network. The performance is compared using multiple robots, and our method is shown to be highly performant on complex (including dual end effector) manipulators.