aiqbench analysis
SAD-Flower addresses critical gaps in flow matching by integrating nonlinear control theory to provide formal guarantees for state and action constraints without requiring retraining. This approach ensures dynamical consistency and admissibility, allowing planners to satisfy unseen constraints at test time while outperforming generative baselines. The work bridges the gap between data-driven planning and rigorous safety requirements essential for real-world robotic systems.
SAD-Flower: Flow Matching for Safe, Admissible, and Dynamically Consistent Planning
arXiv:2511.05355v3 Announce Type: replace Abstract: Flow matching (FM) has shown promising results in data-driven planning. However, it inherently lacks formal guarantees for ensuring state and action constraints, whose satisfaction is a fundamental and crucial requirement for the safety and admissibility of planned trajectories on various systems. Moreover, existing FM planners do not ensure the dynamical consistency, which potentially renders trajectories inexecutable. We address these shortcomings by proposing SAD-Flower, a novel framework for generating Safe, Admissible, and Dynamically consistent trajectories. Our approach relies on an augmentation of the flow with a virtual control input. Thereby, principled guidance can be derived using techniques from nonlinear control theory, providing formal guarantees for state constraints, action constraints, and dynamic consistency. Crucially, SAD-Flower operates without retraining, enabling test-time satisfaction of unseen constraints. Through extensive experiments across several tasks, we demonstrate that SAD-Flower outperforms various generative-model-based baselines in ensuring constraint satisfaction.
Abstract:Flow matching (FM) has shown promising results in data-driven planning. However, it inherently lacks formal guarantees for ensuring state and action constraints, whose satisfaction is a fundamental and crucial requirement for the safety and admissibility of planned trajectories on various systems. Moreover, existing FM planners do not ensure the dynamical consistency, which potentially renders trajectories inexecutable. We address these shortcomings by proposing SAD-Flower, a novel framework for generating Safe, Admissible, and Dynamically consistent trajectories. Our approach relies on an augmentation of the flow with a virtual control input. Thereby, principled guidance can be derived using techniques from nonlinear control theory, providing formal guarantees for state constraints, action constraints, and dynamic consistency. Crucially, SAD-Flower operates without retraining, enabling test-time satisfaction of unseen constraints. Through extensive experiments across several tasks, we demonstrate that SAD-Flower outperforms various generative-model-based baselines in ensuring constraint satisfaction.
Machine Learning (cs.LG); Robotics (cs.RO); Systems and Control (eess.SY)
arXiv:2511.05355 [cs.LG]
(or arXiv:2511.05355v3 [cs.LG] for this version)
Full text under the publisher's open licence · 157 words
Related coverage
Conditional Random Ordered Transport Spaces
arXiv · Jun 9, 2026
Learning Traffic Flow Dynamics with Stochastic Physics-Informed Neural Cellular Automata
arXiv · Oct 8, 2026
Positive Translation-Invariant Solution Found for Cayley Tree Dynamics
Quantum Zeitgeist · Aug 20, 2026
Geometry-Agnostic Data-Driven Discovery of Extreme Scenarios
arXiv · Oct 21, 2026