Recurrent Neural Network Controllers for Signal Temporal Logic Specifications Subject to Safety Constraints

TL;DR

This paper introduces an RNN-based framework for control strategies that satisfy Signal Temporal Logic specifications while ensuring safety via Control Barrier Functions, validated through simulations on safety-critical systems.

Contribution

It presents a novel combination of RNNs and CBFs to generate control policies satisfying STL specifications with safety guarantees.

Findings

RNNs can effectively predict control policies based on system history.

Control Barrier Functions ensure safety of the control policies.

Framework validated through simulations on safety-critical control problems.

Abstract

We propose a framework based on Recurrent Neural Networks (RNNs) to determine an optimal control strategy for a discrete-time system that is required to satisfy specifications given as Signal Temporal Logic (STL) formulae. RNNs can store information of a system over time, thus, enable us to determine satisfaction of the dynamic temporal requirements specified in STL formulae. Given a STL formula, a dataset of satisfying system executions and corresponding control policies, we can use RNNs to predict a control policy at each time based on the current and previous states of system. We use Control Barrier Functions (CBFs) to guarantee the safety of the predicted control policy. We validate our theoretical formulation and demonstrate its performance in an optimal control problem subject to partially unknown safety constraints through simulations.