Worker Support and Training Tools to Aid in Vehicle Quality Inspection for the Automotive Industry

Campaniço, Ana Teresa; Khanal, Salik Ram; Paredes, Hugo; Filipe, Vítor

doi:10.1007/978-3-030-73988-1_35

Cited by 2 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…kSim9 is a multi-purpose simulator, capable of converting the vehicle's digital information into a full 3D visual representation of the correctly assembled vehicle, as well as any other possible combinations. This fully autonomous module was developed as a time and cost saving tool, as it can be used as a training simulator, both for workers [29] and detection algorithms based on simulated data [10], among many other applications. By not depending on still images to form its database, all parts and components can be viewed from any angle, in any lighting condition, making the updating process of both the vehicle models and detection system less time and resource consuming.…”

Section: Vehicle Generatormentioning

confidence: 99%

“…All elements were modeled in Blender 2.8 (https://www.blender.org, accessed on 5 September 2020) and imported to Unity3D (https://unity.com, accessed on 15 November 2021) to build the validation tool used by the factory management and expert observers to confirm the reliability rate of the generated vehicles prior to the development of any software that would use the kSim9 as one of its components. The generator was deemed as an accurate visual representation of the conformity list sample in use, with a reliability rate of 100% [29].…”

Section: Vehicle Generatormentioning

confidence: 99%

See 1 more Smart Citation

Hybrid Quality Inspection for the Automotive Industry: Replacing the Paper-Based Conformity List through Semi-Supervised Object Detection and Simulated Data

et al. 2022

Self Cite

View full text Add to dashboard Cite

The still prevalent use of paper conformity lists in the automotive industry has a serious negative impact on the performance of quality control inspectors. We propose instead a hybrid quality inspection system, where we combine automated detection with human feedback, to increase worker performance by reducing mental and physical fatigue, and the adaptability and responsiveness of the assembly line to change. The system integrates the hierarchical automatic detection of the non-conforming vehicle parts and information visualization on a wearable device to present the results to the factory worker and obtain human confirmation. Besides designing a novel 3D vehicle generator to create a digital representation of the non conformity list and to collect automatically annotated training data, we apply and aggregate in a novel way state-of-the-art domain adaptation and pseudo labeling methods to our real application scenario, in order to bridge the gap between the labeled data generated by the vehicle generator and the real unlabeled data collected on the factory floor. This methodology allows us to obtain, without any manual annotation of the real dataset, an example-based F1 score of 0.565 in an unconstrained scenario and 0.601 in a fixed camera setup (improvements of 11 and 14.6 percentage points, respectively, over a baseline trained with purely simulated data). Feedback obtained from factory workers highlighted the usefulness of the proposed solution, and showed that a truly hybrid assembly line, where machine and human work in symbiosis, increases both efficiency and accuracy in automotive quality control.

show abstract

Section: Vehicle Generatormentioning

confidence: 99%

Section: Vehicle Generatormentioning

confidence: 99%