Safety Critical Intelligent Control of Systems with Uncertain Dynamical Models

Intelligent control has become a cornerstone of modern robotics and autonomous systems, where agents must operate safely and reliably despite uncertainty in their dynamics, environments, and sensing. From autonomous vehicles and aerial drones to robotic manipulators and assistive medical devices, the integration of adaptive, learning-based, and data-driven controllers has expanded the capabilities of these systems to function in highly dynamic contexts. However, achieving formal safety and stability guarantees within such frameworks remains a crucial and unsolved challenge. Uncertainties due to unmodeled dynamics, sensor noise, or real-time learning processes can threaten mission-critical reliability—especially when systems must make autonomous decisions without human oversight.This Research Topic aims to unify advances in safety-critical control theory, optimization-based design, and artificial intelligence to address the central question of how intelligent control methods can be made verifiably safe under uncertainty. The goal is to foster interdisciplinary contributions that build bridges between robust and adaptive control, reinforcement learning, and formal verification approaches. Specifically, this initiative encourages works that demonstrate how safety margins can coexist with learning performance, how controllers can adapt to evolving environments without violating safety constraints, and how theoretical frameworks can support trustworthy autonomous decision-making. By integrating safety certification mechanisms—such as control barrier functions and reachability analysis—with learning architectures, the community can move closer to developing provably safe autonomy for complex robotic systems.To gather further insights into safety-critical intelligent control for uncertain robotic and autonomous systems, we welcome contributions addressing, but not limited to, the following themes:- Learning-based and adaptive control with formal safety guarantees- Safety-critical control under parametric, structural, and environmental uncertainty- Safe reinforcement learning and risk-aware exploration strategies- Control barrier functions, Lyapunov-based methods, and reachability analysis for safety certification- Robust, stochastic, and optimization-driven control for real-world robotics- Fault-tolerant and resilient control architectures under varying conditions- Human-in-the-loop and shared autonomy safety enforcement mechanisms- Distributed and cooperative safety-critical control for multi-agent robotic systems- Data-driven and model-learning approaches ensuring explainable, safe operation- Applications in autonomous vehicles, aerial and underwater robotics, aerospace systems, and industrial automationWe invite original research articles, reviews, perspectives, and methods papers that contribute to theoretical development, algorithmic innovation, and practical experimentation within this Research Topic. Topic editors: Isaac Chairez, Prathyush Menon, Sajjad Keshtkar Source: https://www.frontiersin.org/research-topics/79854/safety-critical-intelligent-control-of-systems-with-uncertain-dynamical-models