The field of control systems is moving towards developing innovative methods for ensuring safety and robustness in complex and dynamic environments. Recent research has focused on creating novel barrier functions, adaptive control algorithms, and data-driven approaches to address the challenges of safety-critical systems. These developments have the potential to significantly improve the performance and reliability of various applications, including robotics, autonomous vehicles, and aerospace systems. Noteworthy papers in this area include: ACORN, which introduces a plug-and-play algorithm for enhancing policy robustness without sacrificing performance. Control Barrier Functions With Real-Time Gaussian Process Modeling, which presents an approach for satisfying state constraints in systems with nonparametric uncertainty. Secure Safety Filter Design for Sampled-data Nonlinear Systems under Sensor Spoofing Attacks, which proposes a secure safety filter design for nonlinear systems under sensor spoofing attacks.
Advancements in Safe and Robust Control of Complex Systems
Sources
Barrier Function Overrides For Non-Convex Fixed Wing Flight Control and Self-Driving Cars
Interval reduced-order switched positive observers for uncertain switched positive linear systems
ACORN: Adaptive Contrastive Optimization for Safe and Robust Fine-Grained Robotic Manipulation
Control Barrier Functions With Real-Time Gaussian Process Modeling
Dynamic Safety in Complex Environments: Synthesizing Safety Filters with Poisson's Equation
Secure Safety Filter Design for Sampled-data Nonlinear Systems under Sensor Spoofing Attacks
Secure Safety Filter: Towards Safe Flight Control under Sensor Attacks
Robust Control of Uncertain Switched Affine Systems via Scenario Optimization
Finite-Sample-Based Reachability for Safe Control with Gaussian Process Dynamics
Non-Conservative Data-driven Safe Control Design for Nonlinear Systems with Polyhedral Safe Sets
Safety and optimality in learning-based control at low computational cost
Feasibility-Aware Pessimistic Estimation: Toward Long-Horizon Safety in Offline RL
Rethink Repeatable Measures of Robot Performance with Statistical Query
Risk-Aware Safe Reinforcement Learning for Control of Stochastic Linear Systems
Automated Statistical Testing and Certification of a Reliable Model-Coupling Server for Scientific Computing
Towards Safe Robot Foundation Models Using Inductive Biases