This paper addresses the critical challenge of vehicle detection in the harsh winter conditions in the Nordic regions, characterized by heavy snowfall, reduced visibility, and low lighting. Due to their susceptibility to environmental distortions and occlusions, traditional vehicle detection methods have struggled in these adverse conditions. The advanced proposed deep learning architectures brought promise, yet the unique difficulties of detecting vehicles in Nordic winters remain inadequately addressed. This study uses the Nordic Vehicle Dataset (NVD), which contains UAV (unmanned aerial vehicle) images from northern Sweden, to evaluate the performance of state-of-the-art vehicle detection algorithms under challenging weather conditions. Our methodology includes a comprehensive evaluation of single-stage, two-stage, segmentation-based, and transformer-based detectors against the NVD. We propose a series of enhancements tailored to each detection framework, including data augmentation, hyperparameter tuning, transfer learning, and Specifically implementing and enhancing the Detection Transformer (DETR). A novel architecture is proposed that leverages self-attention mechanisms with the help of MSER (maximally stable extremal regions) and RST (Rough Set Theory) to identify and filter the region that model long-range dependencies and complex scene contexts. Our findings not only highlight the limitations of current detection systems in the Nordic environment but also offer promising directions for enhancing these algorithms for improved robustness and accuracy in vehicle detection amidst the complexities of winter landscapes. The code and the dataset are available at https://nvd.ltu-ai.dev.

In the pursuit of enhancing smart city infrastructure, computer vision serves as a pivotal element for traffic management, scene understanding, and security applications. This research investigates the performance of the YOLO v9-c and YOLO v9-e object detection models in identifying vehicles under snowy weather conditions, leveraging various data augmentation techniques. The study highlights that, historically, object detection relied on complex, handcrafted features, but deep learning advancements have enabled more efficient and accurate end-to-end learning directly from raw data. Despite these advancements, detecting objects in adverse weather conditions like snow remains challenging, affecting the safety and effectiveness of autonomous systems. The study examines the performance of YOLO v9-c and YOLO v9-e under four different scenarios: no augmentation, snow accumulation, snow overlay, and snow depth mapping. Results indicate that both models achieve their highest precision without augmentation, with YOLO v9-c and YOLO v9-e reaching precisions of 82% and 80%, respectively. However, the snow accumulation method severely impacts detection accuracy, with precision dropping to 36% for YOLO v9-c and 43% for YOLO v9-e. Snow overlay augmentation shows better adaptability, with YOLO v9-c achieving 68% and YOLO v9-e 76% precision. Snow depth mapping results in moderate impacts, with precisions of 59% for YOLO v9-c and 61% for YOLO v9-e. The findings emphasize the importance of careful selection and tuning of augmentation techniques to improve object detection models’ robustness under snowy weather conditions, thereby enhancing the safety and efficiency of autonomous systems. The study suggests a tunned augmentation that helps YOLO v9-c and YOLO v9-e reach precisions of 85% and 83%. Future research should focus more on optimizing augmentation parameters, diversifying training data, and employing domain randomization to further enhance the robustness and generalization capabilities of these models. This approach aims to ensure more reliable performance of autonomous systems in real-world conditions where adverse weather is a common occurrence. The code and the dataset will be available at https://nvd.Itu-ai.dev/

For nearly two decades, business models such as Functional Products have been in focus within research and of interest in the manufacturing industry. Functional product offers consist of hardware, software, service -support systems and management of operation which, when developed in an integrated manner, together provide the customer with an agreed-upon function with a specified level of availability. Compared to product-oriented sales, this type of business model can provide added value to customers, usually through an increase in the service content. Due to the total care commitment, offering Functional Products requires management of reliability and maintainability in order to meet the availability requirement of the function provided. The development of the Functional Product must include holistic analysis and prediction of the functional product availability performance to reduce technical and economic risks and ensure that the function is delivered according to contract. The research performed in this thesis presents an integrated development approach for monitoring and simulation to predict functional product availability. It is shown how the constituents of a functional product can be modelled in an integrated manner in order to simulate and predict functional product availability. A part of this modelling strategy is demonstrated through a simulation case example to show that is possible through this approach to evaluate the availability of different functional product designs. To support the development of the monitoring capability needed for availability simulations it is shown how it is possible to develop fault detection and diagnosis methods for fault detection systems based on data stream management systems. It is also shown how data stream forecasting can be used to predict failures due to faults occurring at short notice. Different fault detection methods have been developed, tested and evaluated on real industrial applications to verify applicability as queries on data streams, managed by data stream management systems. The results from these tests have been evaluated for their predictive performance and detection accuracy. Finally, methodological and technological approaches to monitoring and analysis in functional product development and similar business models to functional products are reviewed. The results showed that few research contributions address the information perspective in functional product development and similar business models holistically. The integrated development approach presented is a pragmatic approach to functional product development which is based on the merged research results of the papers included and knowledge domain presented.

The field of fault detection and diagnosis has been the subject of considerable interest in industry. Fault detection may increase the availability of products, thereby improving their quality. Fault detection and diagnosis methods can be classified in three categories: data-driven, analytically based, and knowledge-based methods. In this work, we investigated the ability and the performance of applying two fault detection methods to query data streams produced from hydraulic drive systems. A knowledge-based method was compared to a data-driven method. A fault detection system based on a data stream management system (DSMS) was developed in order to test and compare the two methods using data from real hydraulic drive systems. The knowledge-based method was based on causal models (fault trees), and principal component analysis (PCA) was used to build the data-driven model. The performance of the methods in terms of accuracy and speed, was examined using normal and physically simulated fault data. The results show that both methods generate queries fast enough to query the data streams online, with a similar level of fault detection accuracy. The industrial applications of both methods include monitoring of individual industrial mechanical systems as well as fleets of such systems. One can conclude that both methods may be used to increase industrial system availability

A functional product is an integrated package consisting of hardware, software and a service support system that provides a customer with a certain function and is sold under a performance-based contract that includes a functional availability guarantee. For the availability performance, prediction, optimisation and management of risk are therefore important concerns during product development. This paper describes a software tool that can generate an integrated model of a functional product from its design details and analyse it through simulation to provide availability performance information. The model's application to the analysis of a real industrial system is demonstrated. Such tools are important for the development and widespread adoption of functional products. The resulting analysis gave an indication of a suitable guaranteed functional availability level for the product and could be used to compare the performance of different design options.

Today, competition among product manufacturers has increased and many facebeing copied and thus losing their uniqueness. To differentiate on the market,new businesses, which extend the manufacturer’s responsibility throughsupporting and guaranteeing product performance, thus offering more value-creating activities, have received increasing interest.The concepts of Functional Products or Total Offers have emerged through the realization that it is tougher to compete with hardware products only. The Functional Products concept combines the development of hardware, software, services and management of operation, thereby integrating value-adding activities into complete offers. Offering a Functional Product is usually a lifecycle commitment (long-term business-to-business agreement) between the function provider and customer. In this type of business, agreed-upon availability is of importance, especially since the availability of a function can directly influence customers’ productivity. The provider and customer need to predict and evaluate the availability of the function, both in early phases of the product development process and continuously during operation throughout the contracted time. In this thesis, in paper A, a model for predicting and monitoring industrial system availability has been proposed to enable the provider of Functional Products to continuously evaluate, through simulation, the offer in terms of availability performance. The model integrates hardware, support system and monitoring system components and through this approach it is possible to utilize operational data in simulations for continuous prediction of availability during operation. Further, the results of paper B presented in this thesis may extend and improve the model proposed in paper A, if implemented. The results of paper B describe the addition of data stream forecasting to the data stream management fault detection system and have showed improved prediction capabilities. The approach to evaluate the availability in paper A may also be extended and improved through the results presented and demonstrated in paper C. In paper C, a software tool to predict system availability has been developed and demonstrated with data from a real industrial system. The results from the simulation demonstrated, together with the addition of the partitioned multi-objective risk method in paper C, provided an indication of what level of availability may be guaranteed for the product; thus, the approach may be used to compare the availability performance of different designs.

This paper describes the integration of a sensor data stream monitoring system into a proposed functional product model capable of predicting functional availability. Such monitoring systems enable predictive maintenance to be carried out pre-emptive maintenance that is scheduled in response to imminent hardware failure and are in widespread use in industry. The industrial motivation for this research is that agreed upon system availability is a critical element of any business-to-business agreement regarding functional sales. Such a model is important when making strategic choices regarding FPs and can be used to develop a high availability product design through simulation driven development, as well as to provide operational decision support that reflects the current reality to enable optimal availability to be achieved in practice. The proposed model integrates hardware, support system and monitoring system models, and is able to incorporate actual operational data. It has been partly verified based on previous research.