safety

3 papers

Safety in robotic manipulation and control.

vla safety foundation-models rl uav manipulation

Papers

Sorted by year (newest first)

Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts

Zhen Sun, Yongjian Guo, Haoran Sun et al. · arXiv preprint · May 2026

A lightweight runtime verification layer that preemptively prunes unsafe VLA and world-model actions before they reach robot hardware, eliminating collision and droppage failures.

vla safety foundation-models

PDF Advanced

No code repo May 2026

Superhuman Safe and Agile Racing through Multi-Agent Reinforcement Learning

Ismail Geles, Leonard Bauersfeld, Markus Wulfmeier et al. · arXiv preprint · May 2026

The first demonstration of superhuman, safe, multi-agent drone racing using a MARL framework trained in simulation and transferred to real Crazyflie nano-drones.

rl uav safety

PDF Advanced

No code repo May 2026

Uncertainty Quantification for Flow-Based Vision-Language-Action Models

Ralf Römer, Maximilian Seeliger, Saida Liu et al. · RSS 2026 — Best Paper Award · Jun 2026

Quantifies epistemic uncertainty in flow-matching VLAs using velocity-field disagreement (VFD) across a small ensemble, enabling failure detection at deployment and sample-efficient active fine-tuning (SAVE).

vla foundation-models safety manipulation

Code PDF Advanced

GitHub ★ — Jun 2026

Suggested Learning Path

Read these papers in order to build expertise in safety.

1
Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts
Zhen Sun, Yongjian Guo et al. · 2026
2
Superhuman Safe and Agile Racing through Multi-Agent Reinforcement Learning
Ismail Geles, Leonard Bauersfeld et al. · 2026
3
Uncertainty Quantification for Flow-Based Vision-Language-Action Models
Ralf Römer, Maximilian Seeliger et al. · 2026