Stochastic System Safety and Control

The field of stochastic systems is moving towards the development of more advanced safety and control methods. Researchers are focusing on creating frameworks that can handle communication imperfections, such as delays and data loss, in networked systems. This involves synthesizing safety controllers that can guarantee system safety in a stochastic sense, ensuring that all trajectories remain within safe regions with a quantified probabilistic bound. Another direction is the construction of barrier certificates, which are being revisited from a dynamic programming perspective to provide tighter probability bounds for safety verification. Additionally, methods for the simultaneous synthesis of barrier certificates and safe controllers for stochastic systems are being developed, allowing for the joint optimization of both within a single formulation. The analysis of rare events in stochastic systems is also an area of interest, with novel heuristics being proposed to efficiently compute lower-bound probabilities of rare events. Finally, control theory-based approaches are being applied to the prevention of diseases at pre-disease states, using techniques such as Markov chain sparse control to identify candidate states for intervention. Noteworthy papers include:

  • A paper that proposes a new set of safety barrier certificate conditions that are strictly less conservative than existing ones, providing tighter probability bounds for safety verification.
  • A paper that presents a method for the simultaneous synthesis of a barrier certificate and a safe controller for discrete-time nonlinear stochastic systems, using piecewise stochastic control barrier functions.
  • A paper that proposes two novel heuristics for priority-first partial state space expansion and trace generation tuned to the transient analysis of rare-event probability in Stochastic Vector Addition Systems.

Sources

Safety Controller Synthesis for Stochastic Networked Systems under Communication Constraints

On the Construction of Barrier Certificate: A Dynamic Programming Perspective

Piecewise Control Barrier Functions for Stochastic Systems

Reasoning about Rare-Event Reachability in Stochastic Vector Addition Systems via Affine Vector Spaces

Designing efficient interventions for pre-disease states using control theory

Built with on top of