Related papers: Runtime Enforcement With Partial Control

On-orbit spacecraft inspection is an important capability for enabling servicing and manufacturing missions and extending the life of spacecraft. However, as space operations become increasingly more common and complex, autonomous control…

In this paper, LLMs are tasked with completing an impossible quiz, while they are in a sandbox, monitored, told about these measures and instructed not to cheat. Some frontier LLMs cheat consistently and attempt to circumvent restrictions…

The performance of reinforcement learning depends upon designing an appropriate action space, where the effect of each action is measurable, yet, granular enough to permit flexible behavior. So far, this process involved non-trivial user…

Monitoring programs for finite state properties is challenging due to high memory and execution time overheads it incurs. Some events if skipped or lost naturally can reduce both overheads, but lead to uncertainty about the current monitor…

This paper introduces NEST (Network-Enforced Session Types), a runtime verification framework that moves application-level protocol monitoring into the network fabric. Unlike prior work that instruments or wraps application code, we…

This paper considers the tracking control problem for an unknown nonlinear system with time-varying bounded disturbance subjected to a prescribed performance and input constraints. When performance and input constraints are specified…

The phenomenon of network vigilantism is autonomously attributed to how anomalies and obscure activities from adversaries can be tracked in realtime. Needless to say, in today dynamic, virtualized, and complex network environments, it has…

We present a design and an implementation of a security policy specification language based on metric linear-time temporal logic (MTL). MTL features temporal operators that are indexed by time intervals, allowing one to specify…

Real-time cyber-physical systems depend on deterministic task execution to guarantee safety and correctness. Unfortunately, this determinism can unintentionally expose timing information that enables adversaries to infer task execution…

Runtime monitors assess whether a system is in an unsafe state based on a stream of observations. We study the problem where the system is subject to probabilistic uncertainty and described by a hidden Markov model. A stream of observations…

Deep reinforcement learning (RL) has shown immense potential for learning to control systems through data alone. However, one challenge deep RL faces is that the full state of the system is often not observable. When this is the case, the…

Robot learning methods have recently made great strides, but generalization and robustness challenges still hinder their widespread deployment. Failing to detect and address potential failures renders state-of-the-art learning systems not…

As factories continue to evolve into collaborative spaces with multiple robots working together with human supervisors in the loop, ensuring safety for all actors involved becomes critical. Currently, laser-based light curtain sensors are…

A robust model predictive control scheme for a class of constrained norm-bounded uncertain discrete-time linear systems is developed under the hypothesis that only partial state measurements are available for feedback. Off-line calculations…

This paper introduces the safety controller architecture as a runtime assurance mechanism for system specifications expressed as safety properties in Linear Temporal Logic (LTL). The safety controller has three fundamental components: a…

LLM-based software engineering agents are increasingly used in real-world development tasks, often with access to sensitive data or security-critical codebases. Such agents could intentionally sabotage these codebases if they were…

Autonomous systems that rely on Machine Learning (ML) utilize online fault tolerance mechanisms, such as runtime monitors, to detect ML prediction errors and maintain safety during operation. However, the lack of human-interpretable…

Monitoring is an important part of the verification toolbox, in particular in situations where exhaustive verification using, e.g., model-checking is infeasible. The goal of online monitoring is to determine the satisfaction or violation of…

Agentic AI systems plan, use tools, maintain state, and produce multi-step trajectories with external effects. Those properties create a governance problem that differs materially from single-turn generative AI: important risks emerge dur-…

When applying imitation learning techniques to fit a policy from expert demonstrations, one can take advantage of prior stability/robustness assumptions on the expert's policy and incorporate such control-theoretic prior knowledge…