Using Petri Nets as an Integrated Constraint Mechanism for Reinforcement Learning Tasks

TL;DR

This paper introduces a method integrating Petri Nets with Reinforcement Learning to enhance trustworthiness, verifiability, and constraint enforcement in control tasks like traffic light management.

Contribution

It presents a novel approach combining Petri Nets with RL, enabling better modeling, constraint enforcement, and property verification for more trustworthy AI systems.

Findings

Outperforms cycle-based baselines in traffic light control

Enables verification of Petri Net properties via model checking

Facilitates combined state modeling with environmental and internal states

Abstract

The lack of trust in algorithms is usually an issue when using Reinforcement Learning (RL) agents for control in real-world domains such as production plants, autonomous vehicles, or traffic-related infrastructure, partly due to the lack of verifiability of the model itself. In such scenarios, Petri nets (PNs) are often available for flowcharts or process steps, as they are versatile and standardized. In order to facilitate integration of RL models and as a step towards increasing AI trustworthiness, we propose an approach that uses PNs with three main advantages over typical RL approaches: Firstly, the agent can now easily be modeled with a combined state including both external environmental observations and agent-specific state information from a given PN. Secondly, we can enforce constraints for state-dependent actions through the inherent PN model. And lastly, we can increase trustworthiness by verifying PN properties through techniques such as model checking. We test our approach on a typical four-way intersection traffic light control setting and present our results, beating cycle-based baselines.