Saved in:
Bibliographic Details
Main Authors: Maus, Tom, Zengeler, Nico, Glasmachers, Tobias
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.10466
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917956088233984
author Maus, Tom
Zengeler, Nico
Glasmachers, Tobias
author_facet Maus, Tom
Zengeler, Nico
Glasmachers, Tobias
contents We present a novel reinforcement learning (RL) environment designed to both optimize industrial sorting systems and study agent behavior in evolving spaces. In simulating material flow within a sorting process our environment follows the idea of a digital twin, with operational parameters like belt speed and occupancy level. To reflect real-world challenges, we integrate common upgrades to industrial setups, like new sensors or advanced machinery. It thus includes two variants: a basic version focusing on discrete belt speed adjustments and an advanced version introducing multiple sorting modes and enhanced material composition observations. We detail the observation spaces, state update mechanisms, and reward functions for both environments. We further evaluate the efficiency of common RL algorithms like Proximal Policy Optimization (PPO), Deep-Q-Networks (DQN), and Advantage Actor Critic (A2C) in comparison to a classical rule-based agent (RBA). This framework not only aids in optimizing industrial processes but also provides a foundation for studying agent behavior and transferability in evolving environments, offering insights into model performance and practical implications for real-world RL applications.
format Preprint
id arxiv_https___arxiv_org_abs_2503_10466
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SortingEnv: An Extendable RL-Environment for an Industrial Sorting Process
Maus, Tom
Zengeler, Nico
Glasmachers, Tobias
Machine Learning
We present a novel reinforcement learning (RL) environment designed to both optimize industrial sorting systems and study agent behavior in evolving spaces. In simulating material flow within a sorting process our environment follows the idea of a digital twin, with operational parameters like belt speed and occupancy level. To reflect real-world challenges, we integrate common upgrades to industrial setups, like new sensors or advanced machinery. It thus includes two variants: a basic version focusing on discrete belt speed adjustments and an advanced version introducing multiple sorting modes and enhanced material composition observations. We detail the observation spaces, state update mechanisms, and reward functions for both environments. We further evaluate the efficiency of common RL algorithms like Proximal Policy Optimization (PPO), Deep-Q-Networks (DQN), and Advantage Actor Critic (A2C) in comparison to a classical rule-based agent (RBA). This framework not only aids in optimizing industrial processes but also provides a foundation for studying agent behavior and transferability in evolving environments, offering insights into model performance and practical implications for real-world RL applications.
title SortingEnv: An Extendable RL-Environment for an Industrial Sorting Process
topic Machine Learning
url https://arxiv.org/abs/2503.10466