Whole-Body Safe Control of Robotic Systems with Koopman Neural Dynamics

Sebin Jung1, Abulikemu Abuduweili1, Jiaxing Li1, Changliu Liu1
1Carnegie Mellon University
Published in 2026 International Conference on Robotics and Automation (ICRA)
Paper

Abstract

Controlling robots with strongly nonlinear, high-dimensional dynamics remains challenging, as direct nonlinear optimization with safety constraints is often intractable in real time. The Koopman operator offers a way to represent nonlinear systems linearly in a lifted space, enabling the use of efficient linear control. We propose a data-driven framework that learns a Koopman embedding and operator from data, and integrates the resulting linear model with the Safe Set Algorithm (SSA). This allows the tracking and safety constraints to be solved in a single quadratic program (QP), ensuring feasibility and optimality without a separate safety filter. We validate the method on a Kinova Gen3 manipulator and a Go2 quadruped, showing accurate tracking and obstacle avoidance.

Whole-Body Collision Avoidance Performance

Generalizable to Various Robots

Kinova Gen3 - 7DOF

Kinova Gen3 - 6DOF

Kinova Gen3 Lite

Kuka iiwa7

Franka Emika Panda

Unitree Z1

Ufactory Lite6

Extension to Floating-base Systems

Learning Whole-Body Dynamics with Neural Lifting

Reference Trajectory to Track

Safe Control with KMPC

Pipeline Overview

Pipeline overview diagram

BibTeX


        @misc{jung2026wholebodysafecontrolrobotic,
          title={Whole-Body Safe Control of Robotic Systems with Koopman Neural Dynamics}, 
          author={Sebin Jung and Abulikemu Abuduweili and Jiaxing Li and Changliu Liu},
          year={2026},
          eprint={2603.03740},
          archivePrefix={arXiv},
          primaryClass={cs.RO},
          url={https://arxiv.org/abs/2603.03740}, 
    }