Hartmann A, Liu Z, Lamprecht S, Bründl P, Franke J (2026)
Publication Type: Conference contribution
Publication year: 2026
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 766 IFIPAICT
Pages Range: 349-363
Conference Proceedings Title: IFIP Advances in Information and Communication Technology
Event location: Kamakura, JPN
ISBN: 9783032035370
DOI: 10.1007/978-3-032-03538-7_25
The rapid evolution of electric, connected, autonomous, and shared (ECAS) vehicles is transforming automotive architectures and driving demand for complex, scalable, and efficient wiring systems, even as harness assembly remains a predominantly manual and ergonomically challenging process. However, reliable verification of electrical connector mating, traditionally performed by human operators via auditory and tactile feedback, remains challenging for full automation. This study introduces a multimodal sensing approach integrating acoustic, force, and kinematic data for robust, real-time detection of successful electrical connector mating. High-frequency acoustic signals capturing mechanical click signatures, combined with simultaneous force sensor data and robot end-effector motion profiles, provide complementary information to resolve ambiguous events. Various supervised learning algorithms, including convolutional neural networks (CNNs), multilayer perceptrons (MLPs), and random forest classifiers, are evaluated using a dataset including diverse connector types and ambient noise levels. Feature extraction techniques and dynamic thresholding mechanisms isolate critical signal features, enhancing performance under low signal-to-noise conditions. Designed for seamless robotic integration, the system delivers immediate feedback for downstream assembly processes. Achieving up to 96.8% accuracy with the deployed CNN, the approach demonstrates the viability of AI-driven multisensor fusion for reliable connector verification, facilitating agile, sustainable, and digitally integrated manufacturing systems.
APA:
Hartmann, A., Liu, Z., Lamprecht, S., Bründl, P., & Franke, J. (2026). AI-Driven Multisensor Quality Inspection: A Focus on Robotic Wire Harness Assembly. In Hajime Mizuyama, Eiji Morinaga, Tomomi Nonaka, Toshiya Kaihara, Gregor von Cieminski, David Romero (Eds.), IFIP Advances in Information and Communication Technology (pp. 349-363). Kamakura, JPN: Springer Science and Business Media Deutschland GmbH.
MLA:
Hartmann, Annalena, et al. "AI-Driven Multisensor Quality Inspection: A Focus on Robotic Wire Harness Assembly." Proceedings of the 44th IFIP WG 5.7 International Conference on Advances in Production Management Systems, APMS 2025, Kamakura, JPN Ed. Hajime Mizuyama, Eiji Morinaga, Tomomi Nonaka, Toshiya Kaihara, Gregor von Cieminski, David Romero, Springer Science and Business Media Deutschland GmbH, 2026. 349-363.
BibTeX: Download