Search

Autonomous driving poses numerous challenging problems, one of which is perceiving and understanding the environment. Since self-driving is safety critical and many actions taken during driving rely on the outcome of various perception algorithms (for instance all traffic participants and infrastructural objects in the vehicle's surroundings must reliably be recognized and localized), thus the perception might be considered as one of the most critical subsystems in an autonomous vehicle. Although the perception itself might further be decomposed into various sub-problems, such as object detection, lane detection, traffic sign detection, environment modeling, etc. In this paper the focus is on fusion models in general (giving support for multisensory data processing) and some related automotive applications such as object detection, traffic sign recognition, end-to-end driving models and an example of taking decisions in multi-criterial traffic situations that are complex for both human drivers and for the self-driving vehicles as well.

We demonstrate a working functional prototype of a cooperative perception system that maintains a real-time digital twin of the traffic environment, providing a more accurate and more reliable model than any of the participant subsystems—in this case, smart vehicles and infrastructure stations—would manage individually. The importance of such technology is that it can facilitate a spectrum of new derivative services, including cloud-assisted and cloud-controlled ADAS functions, dynamic map generation with analytics for traffic control and road infrastructure monitoring, a digital framework for operating vehicle testing grounds, logistics facilities, etc. In this paper, we constrain our discussion on the viability of the core concept and implement a system that provides a single service: the live visualization of our digital twin in a 3D simulation, which instantly and reliably matches the state of the real-world environment and showcases the advantages of real-time fusion of sensory data from various traffic participants. We envision this prototype system as part of a larger network of local information processing and integration nodes, i.e., the logically centralized digital twin is maintained in a physically distributed edge cloud.

The paper discusses the identification of the empirical white noise processes generated by deterministic numerical algorithms.The introduced fuzzy-random complementary approach can identify the inner hidden correlational patterns of the empirical white noise process if the process has a real hidden structure of this kind. We have shown how the characteristics of auto-correlated white noise processes change as the order of autocorrelation increases. Although in this paper we rely on random number generators to get approximate white noise processes, in our upcoming research we are planning to turn the focus on physical white noise processes in order to validate our hypothesis.

A spectacular measurement campaign was carried out on a real-world motorway stretch of Hungary with the participation of international industrial and academic partners. The measurement resulted in vehicle based and infrastructure based sensor data that will be extremely useful for future automotive R&D activities due to the available ground truth for static and dynamic content. The aim of the measurement campaign was twofold. On the one hand, road geometry was mapped with high precision in order to build Ultra High Definition (UHD) map of the test road. On the other hand, the vehicles—equipped with differential Global Navigation Satellite Systems (GNSS) for ground truth localization—carried out special test scenarios while collecting detailed data using different sensors. All of the test runs were recorded by both vehicles and infrastructure. The paper also showcases application examples to demonstrate the viability of the collected data having access to the ground truth labeling. This data set may support a large variety of solutions, for the test and validation of different kinds of approaches and techniques. As a complementary task, the available 5G network was monitored and tested under different radio conditions to investigate the latency results for different measurement scenarios. A part of the measured data has been shared openly, such that interested automotive and academic parties may use it for their own purposes.

The paper discusses the identification of the empirical, partially intelligible white noise processes generated by deterministic numerical algorithms. The introduced fuzzy‐random complemen...

Autonomous driving functions increasingly incorporate machine learning methods for perception, environment modeling and decision making. The ever-present danger of unintended behavior of neural networks has given cause for concern in safety-critical applications and has spurred research towards the establishment of certified properties. In our work we present a general approach for exploring guaranteed properties of neural networks through a specific use case. We first methodically specified several desired properties for an ideal traffic sign classifier. Then we applied a mixed integer linear programming verification method to certify (or conversely, falsify) that our pre-trained rectified linear unit (ReLU) activation network will satisfy given properties under all conditions. We established through relaxations of our initial specification that our model does in fact happen to guarantee certain (but not all) properties at a different threshold. We thus showed that it is possible to establish the degree to which non-trivial and humanly meaningful properties are inherently guaranteed by an arbitrary pixel space ReLU network. We believe this approach can be a helpful addition in comprehensively evaluating the reliability of a network via quantifying its degree of compliance to an ideal specification.

In real time evaluation systems the decision should be made in time to avoid the serious consequences. In the literature there are several techniques which aim to reduce the computational complexity of these kinds of systems. Among others the Higher Order Singular Value Decomposition (HOSVD) based reduction method is a useful tool for handling this problem. In the case when only non-exact reduction can be performed, the error calculation is also essential, because the reduction degree is based on the calculated error in these systems. In this paper the authors present the HOSVD reduction error calculation in hierarchical fuzzy systems using Mamdani-type inference and define a general formula which includes the propagated error from the previous levels of the hierarchy; in which the error bound, calculated by different ways, can be used and takes into account the case when not all the inputs contain a propagated error but only a few of them. Furthermore a greedy algorithm is presented for reduction optimization.

The paper focuses on the problem of raw data fusion in neural networks based 3D object detection architectures. Here we consider the case of autonomous driving with data from camera and LiDAR sensors. Understanding the vehicle surroundings is a crucial task in autonomous driving since any subsequent action taken is strongly dependent on it. In this paper we present an alternative method of fusing camera image information with LiDAR poinclouds at a close to raw level of abstraction. Our results suggest that our approach improves the average precision of 3D bounding box detection of cyclists (and possibly other objects) in sparse point clouds compared to the baseline architecture without low-level fusion. The proposed approach has been evaluated on the KITTI dataset containing driving scenes with corresponding camera and LiDAR data. The long-term goal of our research is to develop a neural network architecture for environment perception that fuses multi-sensor data at the earliest stages possible, thus leveraging the full benefits of possible inter-sensor synergies.

This paper describes the relation of the higher order singular value decomposition (HOSVD) based canonical form to the Hilbert-Schmidt type integral operators. This connection between the canonical form of f(x) and the integral operators determined by function f(x) is important; on the one hand it gives a new aspect for the consideration of LPV models, and on the other hand it allows for the approximation of eigenfunctions and integral operators based on a numerical reconstruction of the canonical form.

Photo sharing websites have become extremely popular with GPS-enabled digital cameras and camera phones. Seeing autumn leaves, cherry blossoms, etc. is a traditional seasonal activity. Tremendous photos about leaves and blossoms have been uploaded. The goal of our paper is to detect and rank the most scenic leaves and blossoms viewing places based on social and image features. Methods for social interestingness, color percentage, edge rate, sharpness, ranking score have been provided. The proposed methods have been validated and verified by experimental results for maple leaves in Kyoto.

To reveal the influence of global warming on glaciers, highly accurate observations of glacier movement must continue every year. It is thought that there is a close relationship between glacier moving speed and global warming. Thus, there have been precise, detailed observations of the movement of the Perito Moreno glacier in Patagonia of the Argentine Republic over the past five years. The measurement method of using GPS and an optical measuring instrument is generally used to monitor glacier movement, but the measurement accuracy attained is not optimal because of the huge size of the glacier. The measurement system used for the Perito Moreno observations, however, could realize high accuracy measurement over long distance. The measurement system is based on the principle of stereo measurement using cameras. This paper describes the results of the Perito Moreno glacier observations of this year and considers the effectiveness of glacier observation using cameras.

A time series collected by thermal cameras yields many application possibilities for security monitoring applications. The character of features involved in thermal images is different compared to the images acquired by conventional capturing devices; many methods applicable for conventional image processing cannot directly be applied for thermal images. New methods are introduced in HOSVD based representation of thermal images, and in application of resolution enhancement and data compression. The retrieval process of thermal image databases and useful feature descriptors were analyzed, as well as the special semi-automatic fusion techniques of thermal image sequences.

With GPS-enabled digital cameras and camera phones photo sharing websites have become exceptionally popular recently. Traditional seasonal activities are to admire leaves and blossoms. Tremendous photos about them have been uploaded. The goal of our paper is to detect the most scenic leaves and blossoms viewing places from Flickr based on social and image features as well as to perform their time trend analysis based on statistical and control engineering methods. Firstly, methods for social interestingness, color percentage, edge rate, sharpness, ranking score have been proposed. Secondly, methods for time trend analysis based on chi square test of independence and subspace identification have been provided. The proposed methods have been validated and verified with experimental results for maple leaves in Kyoto.

High dynamic range of illumination may cause serious distortions and other problems in viewing and further processing of digital images. In this paper a new fuzzy based tone reproduction pre-processing algorithm is introduced which may help in developing hardly or nonviewable features and content of the images making easier the further processing of it.

In the paper the authors proposed a HOSVD based approach to enhance the reconstruction of inner structures based on projections from different angles which is strongly related to the well known Radon transform. It is shown that in case of less number of projections the HOSVD based approximation applied in the projection space can significantly enhance the output obtained by the inverse Radon transform.

The representation of data plays significant role in many applications, as for instance when performing data compression, feature extraction or enhancement, etc. In this paper we briefly mention some well known data representation forms and propose a new domain based on the so called higher order singular value decomposition (HOSVD) and wavelet transformation. It will be shown how the data can be processed by manipulating its components in this domain. Furthermore, the properties of the components as well as the applicability of the proposed approach in the field of image processing and system identification will be shown.

The paper describes an approach for data representation and processing in the so called HOSVD-DCT based domain which combines the higher order singular value decomposition (HOSVD) with the discrete cosine transformation (DCT). The key idea is to express the multivariate function by an orthonormal set of one variable specific functions and perform DCT on those components which are less significant from the application point of view in order to ensure finer grained processing. It is advantageous also in case of system identification, where the behaviour of the system can be approximated by blending locally identified models. In this case the proposed method can efficiently be applied.

The paper deals with the theoretical background of the higher order singular value decomposition (HOSVD) based canonical form of functions. Furthermore in special case it describes the relation between the canonical form and the Hilbert-Schmidt type integral operators. The described techniques have a variety of applications, e.g. image processing, system identification, data compression, filtering, etc. As an example of application from the field of intelligent systems, a tensor-product based concept is introduced useful for approximating the behavior of a strongly non-linear system by locally tuned neural network models. The proposed approach may be a useful tool for solving many kind of black-box like identification problems. The weights in the corresponding layers of the input local models are jointly expressed in tensor-product form such a way ensuring the efficient approximation. Similar concept has been used by the authors for approximating the system matrix of linear parameter varying systems in state space representation. We hope that the proposed concept could be an efficient compromised modeling view using both the analytical and heuristic approaches.

The spread of Web 2.0 has caused user-generated content explosion. Users can tag resources in order to describe and organize them. Tag clouds are visually depicted tags in order to facilitate browsing among numerous tags and resources. The goal of our paper is to improve tag clouds with vocabulary refinement and enhanced reference counts. Namely, three novel algorithms have been proposed to correct spelling and clerical errors, contract tags, and improve reference counts. The provided algorithms have been validated and verified on real-world tag clouds of a thesis portal.

Data compression and enhancement represent an important consideration in many application areas. The transmission bandwidth and storage capacity are often crucial factors. The efficiency of the processing of data in numerous cases strongly depends on the form of data representation. In the present paper a method, based on the so called higher order singular value decomposition and tensor-product transformation, is introduced for multidimensional scaling and compression of thermal images sequences. The proposed approach operates with smooth functions forming an orthonormal basis. Because of the smoothness property of these orthonormal components, the method can advantageously be utilized for the efficient compression of thermal image sequences.

Anytime systems represent a subfamily of the realtime ones. They provide continuous operation even in cases when the circumstances are changing, e.g temporary shortage of time, changing computational power, etc. As the available computation time decreases, the accuracy of the computation becomes negatively affected. Iterative algorithms support this kind of computation. Unfortunately the construction of iterative algorithms is complex and their usability is limited. In this paper a novel method for anytime systems with modular architecture is introduced. The main aim of the proposed algorithm is to extend the applicability of anytime systems. The effectiveness of the algorithm is demonstrated on a categorization problem.

The approximation methods of mathematics are widely used in theory and practice for several problems. In the framework of the paper a novel tensor-product based approach for representation of neural networks (NNs) is proposed. The NNs in this case stand for local models based on which a more complex parameter varying model can numerically be reconstructed and reduced using the higher order singular value decomposition (HOSVD). The HOSVD as well as the tensor-product based representation of NNs will be discussed in detail.

In this paper we present a method for constructing a control model of loading systems based on higher order singular decomposition (HOSVD) of a tensor. Modern logistics need control systems that are able to develop and improve the material and information flow by automating the control processes. We propose the application of linear parameter varying (LPV) structure by which non-linear systems can be controlled on the basis of linear control theories. Describing the system, modeling logistic processes require many uncertain input parameters. Using the proposed method the complexity of the model can be kept on a lower level.

Process chains are a common modeling paradigm for analysis and optimization of logistic processes, and are intensively used in many practical applications. The ProC/B toolset is a collection of software tools for modeling, analysis, validation and optimization of process chains. The ProC/B models can be translated into queueing networks or Petri nets, which can be solved by effective techniques and algorithms to evaluate performance metrics. The base queueing model with Mean-Value Analysis evaluation algorithm, and their adaptations for modeling thread pool and queue limit have been verified and validated for multi-tier software systems. The goal of our work is to adapt these models and algorithms for process chains to model parallel processes and queue limit.