Deep Reinforcement Learning Based Networked Control with Network Delays for Signal Temporal Logic Specifications

TL;DR

This paper introduces a deep reinforcement learning approach for designing networked controllers that account for network delays and satisfy signal temporal logic specifications in dynamical systems.

Contribution

It extends the Markov decision process to include past states and actions to handle network delays, enabling effective DRL-based control for STL specifications.

Findings

Successful application of DRL to networked control with delays

Effective satisfaction of STL specifications in simulations

Demonstrated learning performance of the proposed method

Abstract

We apply deep reinforcement learning (DRL) to design of a networked controller with network delays to complete a temporal control task that is described by a signal temporal logic (STL) formula. STL is useful to deal with a specification with a bounded time interval for a dynamical system. In general, an agent needs not only the current system state but also the past behavior of the system to determine a desired control action for satisfying the given STL formula. Additionally, we need to consider the effect of network delays for data transmissions. Thus, we propose an extended Markov decision process using past system states and control actions, which is called a $\tau d$-MDP, so that the agent can evaluate the satisfaction of the STL formula considering the network delays. Thereafter, we apply a DRL algorithm to design a networked controller using the $\tau d$-MDP. Through simulations, we also demonstrate the learning performance of the proposed algorithm.