Abstract: We propose a Dynamical System (DS) approach to learn complex, possibly periodic motion plans from kinesthetic demonstrations using Neural Ordinary Differential Equations (NODE). To ensure reactivity and robustness to disturbances, we propose a novel approach that selects a target point at each time step for the robot to follow, by combining tools from control theory and the target trajectory generated by the learned NODE.  A correction term to the NODE model is computed online by solving a quadratic program that guarantees stability and safety using Control Lyapunov Functions (CLFs) and Control Barrier Functions (CBFs), respectively. Our approach outperforms baseline DS learning techniques on the LASA handwriting dataset and complex periodic trajectories. It is also validated on the Franka Emika robot arm to produce stable motions for wiping and stirring tasks that do not have a single attractor, while being robust to perturbations and safe around humans and obstacles.

Demonstrations from Franka robot and corresponding NODE predictions: {1, 2, 3} denote each column of the rotation matrix.

NODE model for predicting full pose

Pouring

Plate stacking

Box opening

Bottle shelving

Illustration of a spurious attractor

Disturbance rejection and obstacle avoidance on LASA data set