The advent of Industry 4.0 has shown the tremendous transformative potential of combining artificial intelligence, cyber-physical systems and Internet of Things concepts in industrial settings. Despite this, data availability is still a major roadblock for the successful adoption of data-driven solutions, particularly concerning deep learning approaches in manufacturing. Specifically in the quality control domain, annotated defect data can often be costly, time-consuming and inefficient to obtain, potentially compromising the viability of deep learning approaches due to data scarcity. In this context, we propose a novel method for generating annotated synthetic training data for automated quality inspections of structural adhesive applications, validated in an industrial cell for automotive parts. Our approach greatly reduces the cost of training deep learning models for this task, while simultaneously improving their performance in a scarce manufacturing data context with imbalanced training sets by 3.1% (mAP@0.50). Additional results can be seen at https://ricardosperes.github.io/simulation-synth-adhesive/.

The technological advances brought forth by the Industry 4.0 paradigm have renewed the disruptive potential of artificial intelligence in the manufacturing sector, building the data-driven era on top of concepts such as Cyber–Physical Systems and the Internet of Things. However, data availability remains a major challenge for the success of these solutions, particularly concerning those based on deep learning approaches. Specifically in the quality inspection of structural adhesive applications, found commonly in the automotive domain, defect data with sufficient variety, volume and quality is generally costly, time-consuming and inefficient to obtain, jeopardizing the viability of such approaches due to data scarcity. To mitigate this, we propose a novel approach to generate synthetic training data for this application, leveraging recent breakthroughs in training generative adversarial networks with limited data to improve the performance of automated inspection methods based on deep learning, especially for imbalanced datasets. Preliminary results in a real automotive pilot cell show promise in this direction, with the approach being able to generate realistic adhesive bead images and consequently object detection models showing improved mean average precision at different thresholds when trained on the augmented dataset. For reproducibility purposes, the model weights, configurations and data encompassed in this study are made publicly available.

Industrial environments are heterogeneous systems that create challenges of interoperability limiting the development of systems capable of working collaboratively from the point of view of machines and software. Additionally, environmental issues related to manufacturing systems have emerged during the last decades, related to sustainability problems faced in the world. Thus, the proposed work aims to present an interoperable solution based on events to reduce the complexity of integration, while creating energetic profiles for the machines to allow the optimization of their energy consumption. A publish/subscribe-based architecture is proposed, where the instantiation is based on Apache Kafka. The proposed solution was implemented in two robotic cells in the automotive industry, constituted by different hardware, which allowed testing the integration of different components. The energy consumption data was then sent to a Postgres database where a graphical interface allowed the operator to monitor the performance of each cell regarding energy consumption. The results are promising due to the system’s ability to integrate tools from different vendors and different technologies. Furthermore, it allows the possibility to use these developments to deliver more sustainable systems using more advanced solutions, such as production scheduling, to reduce energy consumption.

This paper presents a robotic team suited for bottom sediment sampling and retrieval in mudflats, targeting environmental monitoring tasks. The robotic team encompasses a four-wheel-steering ground vehicle, equipped with a drilling tool designed to be able to retain wet soil, and a multi-rotor aerial vehicle for dynamic aerial imagery acquisition. On-demand aerial imagery, properly fused on an aerial mosaic, is used by remote human operators for specifying the robotic mission and supervising its execution. This is crucial for the success of an environmental monitoring study, as often it depends on human expertise to ensure the statistical significance and accuracy of the sampling procedures. Although the literature is rich on environmental monitoring sampling procedures, in mudflats, there is a gap as regards including robotic elements. This paper closes this gap by also proposing a preliminary experimental protocol tailored to exploit the capabilities offered by the robotic system. Field trials in the south bank of the river Tagus’ estuary show the ability of the robotic system to successfully extract and transport bottom sediment samples for offline analysis. The results also show the efficiency of the extraction and the benefits when compared to (conventional) human-based sampling.

Stereo-based off-road obstacle detection is a complex and still open problem. The challenges are in great extent related to computational cost and noise level. Previous work has shown that visual saliency and voting mechanisms are extremely effective in tackling these issues. This paper proposes a set of extensions to these mechanisms, to further improve the detector's speed-accuracy trade-off as well as its robustness. The observed enhancements are due in part to the adaptive way saliency is accounted for by the detector during the image scanning procedure. Additionally, detector's positive results are in turn used to boost the saliency map itself, thus reinforcing the analysis of relevant regions of the image. To enable detection in highly roughed terrain, the detector's invariance to the robot's posture is also enhanced. Experimental results show that, with the extended detector, denominated of ESalOD, higher levels of robustness, accuracy and computational efficiency are attained.