Evaluation of a Smart Mobile Robotic System for Industrial Plant Inspection and Supervision

TL;DR

The paper presents a multimodal autonomous mobile robotic system for industrial plant inspection that combines olfactory, optical, and acoustic sensing with autonomous navigation. It demonstrates integrated sensing (electronic nose, active IR gas camera, UV camera, microphone array, LiDAR mapping, and passive cameras) and a ROS-driven software stack enabling 2D/object detection, 3D localization, mapping, and change detection, validated in a real chemical plant. Key contributions include real-time visualization and localization of methane leaks down to 40 mL/min, robust acoustic leak localization with sub-degree DoA accuracy, and a data-driven anomaly detection approach in industrial acoustics; along with a comprehensive evaluation of navigation, gas sensing, and acoustic capabilities. The work advances practical autonomous plant supervision by validating multimodal sensing and navigation in a challenging wastewater-treatment environment, highlighting pathways toward scalable, 24/7 robotic inspection in industry.

Abstract

Automated and autonomous industrial inspection is a longstanding research field, driven by the necessity to enhance safety and efficiency within industrial settings. In addressing this need, we introduce an autonomously navigating robotic system designed for comprehensive plant inspection. This innovative system comprises a robotic platform equipped with a diverse array of sensors integrated to facilitate the detection of various process and infrastructure parameters. These sensors encompass optical (LiDAR, Stereo, UV/IR/RGB cameras), olfactory (electronic nose), and acoustic (microphone array) capabilities, enabling the identification of factors such as methane leaks, flow rates, and infrastructural anomalies. The proposed system underwent individual evaluation at a wastewater treatment site within a chemical plant, providing a practical and challenging environment for testing. The evaluation process encompassed key aspects such as object detection, 3D localization, and path planning. Furthermore, specific evaluations were conducted for optical methane leak detection and localization, as well as acoustic assessments focusing on pump equipment and gas leak localization.

Figures (13)

• Figure 1: The mobile robotic platform with multimodal sensors.
• Figure 2: System architecture with communication between the components. Solid or dotted lines between the boxes represent, respectively, Ethernet or wireless connectivity.
• Figure 3: Setup of the active gas camera. A detailed description of its working principle can be found in bergau_real-time_2023.
• Figure 4: UCA Microphone Array with a windscreen for outdoor usage. Details of the employed algorithms can be found in Fischer.2021
• Figure 5: Sequential steps of the DoA estimation algorithm, starting with a coarse but computationally efficient guess and progressing to a more detailed peak search in the wideband MUSIC spectrum for refined accuracy.