Abstract:Several methods have been proposed recently to learn neural network (NN) controllers for autonomous agents, with unknown and stochastic dynamics, tasked with complex missions captured by Linear Temporal Logic (LTL). Due to the sample-inefficiency of the majority of these works, compositional learning methods have been proposed decomposing the LTL specification into smaller sub-tasks. Then, separate controllers are learned and composed to satisfy the original task. A key challenge within these approaches is that they often lack safety guarantees or the provided guarantees are impractical. This paper aims to address this challenge. Particularly, we consider autonomous systems with unknown and stochastic dynamics and LTL-encoded tasks. We assume that the system is equipped with a finite set of base skills modeled by trained NN feedback controllers. Our goal is to check if there exists a temporal composition of the trained NN controllers - and if so, to compute it - that will yield a composite system behavior that satisfies the assigned LTL task with probability one. We propose a new approach that relies on a novel integration of automata theory and data-driven reachability analysis tools for NN-controlled stochastic systems. The resulting neuro-symbolic controller allows the agent to generate safe behaviors for unseen complex temporal logic tasks in a zero-shot fashion by leveraging its base skills. We show correctness of the proposed method and we provide conditions under which it is complete. To the best of our knowledge, this is the first work that designs verified temporal compositions of NN controllers for unknown and stochastic systems. Finally, we provide extensive numerical simulations and hardware experiments on robot navigation tasks to demonstrate the proposed method.

Leveraging Weighted Automata in Compositional Reasoning about Concurrent Probabilistic Systems

Learning Weighted Assumptions for Compositional Verification of Markov Decision Processes

Should We Learn Probabilistic Models for Model Checking? A New Approach and An Empirical Study.

Compositional Reasoning for Shared-variable Concurrent Programs.

Verified Compositional Neuro-Symbolic Control for Stochastic Systems with Temporal Logic Tasks

Deterministic Weighted Automata under Partial Observability

On Compositional Reasoning for Guaranteeing Probabilistic Properties

Compositional Verification for Autonomous Systems with Deep Learning Components

Abstraction and Learning for Infinite-State Compositional Verification

Cost Preserving Bisimulations for Probabilistic Automata

Safety-Aware Task Composition for Discrete and Continuous Reinforcement Learning

Mixed Semantics Guided Layered Bounded Reachability Analysis of Compositional Linear Hybrid Automata

Automatic compositional reasoning for multi-thread programs

Compositional Model-Checking for Real-Time Systems

A unified concurrent-composition method to state/event inference and concealment in discrete-event systems

Compositional Stochastic Modeling and Probabilistic Programming

Compositional Design of Safety Controllers for Large-scale Stochastic Hybrid Systems

From Small-Gain Theory to Compositional Construction of Barrier Certificates for Large-Scale Stochastic Systems

Compositional Abstractions of Interconnected Discrete-Time Stochastic Control Systems

Reachability Assume-Guarantee

Assumption Generation for the Verification of Learning-Enabled Autonomous Systems