SafeFlowMPC: Predictive and Safe Trajectory Planning for Robot Manipulators with Learning-based Policies

TL;DR

SafeFlowMPC is a novel approach that combines learning-based policies with optimization techniques to ensure safe, real-time trajectory planning for robot manipulators, demonstrated on real-world tasks with a KUKA robot.

Contribution

It introduces SafeFlowMPC, a hybrid method that guarantees safety and real-time performance by integrating flow matching with online optimization for robot trajectory planning.

Findings

Successfully applied to grasping and human-robot handover tasks

Ensures safety at all times during manipulation

Achieves real-time performance with suboptimal model predictive control

Abstract

The emerging integration of robots into everyday life brings several major challenges. Compared to classical industrial applications, more flexibility is needed in combination with real-time reactivity. Learning-based methods can train powerful policies based on demonstrated trajectories, such that the robot generalizes a task to similar situations. However, these black-box models lack interpretability and rigorous safety guarantees. Optimization-based methods provide these guarantees but lack the required flexibility and generalization capabilities. This work proposes SafeFlowMPC, a combination of flow matching and online optimization to combine the strengths of learning and optimization. This method guarantees safety at all times and is designed to meet the demands of real-time execution by using a suboptimal model-predictive control formulation. SafeFlowMPC achieves strong performance in three real-world experiments on a KUKA 7-DoF manipulator, namely two grasping experiment and a dynamic human-robot object handover experiment. A video of the experiments is available at http://www.acin.tuwien.ac.at/42d6. The code is available at https://github.com/TU-Wien-ACIN-CDS/SafeFlowMPC.