Search Results

Abstract Lane changes are stressful maneuvers for drivers, particularly during high-speed traffic flows. However, modeling driver’s lane-changing decision and implementation process is challenging due to the complexity and uncertainty of driving behaviors. To address this issue, this article presents a personalized Lane-Changing Model (LCM) for Advanced Driver Assistance System (ADAS) based on deep learning method. The LCM contains three major computational components. Firstly, with abundant inputs of Root Residual Network (Root-ResNet), LCM is able to exploit more local information from the front view video data. Secondly, the LCM has an ability of learning the global spatial-temporal information via Temporal Modeling Blocks (TMBs). Finally, a two-layer Long Short-Term Memory (LSTM) network is used to learn video contextual features combined with lane boundary based distance features in lane change events.

Abstract Real-time reconstruction of 3D environment attributed with semantic information is significant for a variety of applications, such as obstacle detection, traffic scene comprehension and autonomous navigation. The current approaches to achieve it are mainly using stereo vision, Structure from Motion (SfM) or mobile LiDAR sensors. Each of these approaches has its own limitation, stereo vision has high computational cost, SfM needs accurate calibration between a sequences of images, and the onboard LiDAR sensor can only provide sparse points without color information. This paper describes a novel method for traffic scene semantic segmentation by combining sparse LiDAR point cloud (e.g. from Velodyne scans), with monocular color image. The key novelty of the method is the semantic coupling of stereoscopic point cloud with color lattice from camera image labelled through a Convolutional Neural Network (CNN).

Abstract The innovations of vehicle connectivity have been increasing dramatically to enhance the safety and user experience of driving, while the rising numbers of interfaces to the external world also bring security threats to vehicles. Many security countermeasures have been proposed and discussed to protect the systems and services against attacks. To provide an overview of the current states in this research field, we conducted a systematic mapping study (SMS) on the topic area “security countermeasures of in-vehicle communication systems.” A total of 279 papers are identified based on the defined study identification strategy and criteria. We discussed four research questions (RQs) related to the security countermeasures, validation methods, publication patterns, and research trends and gaps based on the extracted and classified data. Finally, we evaluated the validity threats and the whole mapping process.

Abstract This work introduces the concept of a dual-layer specification structure for standards that separate interoperability functions, such as backward compatibility, localization, and deployment, from those essential to reliability, security, and functionality. The latter group of features, which constitute the actual standard, make up the baseline layer for instructions, while all the elements required for interoperability are specified in a second layer, known as a Protocols, Operations, Usage, and Formats (POUF) document. We applied this technique in the development of a standard for Uptane [1], a security framework for over-the-air (OTA) software updates used in many automobiles. This standard is a good candidate for a dual-layer specification because it requires communication between entities, but does not require a specific format for this communication.

Abstract Modern vehicles integrate Internet of Things (IoT) components to bring value-added services to both drivers and passengers. These components communicate with the external world through different types of interfaces including the on-board diagnostics (OBD-II) port, a mandatory interface in all vehicles in the United States and Europe. While this transformation has driven significant advancements in efficiency and safety, it has also opened a door to a wide variety of cyberattacks, as the architectures of vehicles were never designed with external connectivity in mind, and accordingly, security has never been pivotal in the design. As standardized, the OBD-II port allows not only direct access to the internal network of the vehicle but also installing software on the Electronic Control Units (ECUs).

Abstract In the automotive domain, the overall complexity of technical components has increased enormously. Formerly isolated, purely mechanical cars are now a multitude of cyber-physical systems that are continuously interacting with other IT systems, for example, with the smartphone of their driver or the backend servers of the car manufacturer. This has huge security implications as demonstrated by several recent research papers that document attacks endangering the safety of the car. However, there is, to the best of our knowledge, no holistic overview or structured description of the complex automotive domain. Without such a big picture, distinct security research remains isolated and is lacking interconnections between the different subsystems. Hence, it is difficult to draw conclusions about the overall security of a car or to identify aspects that have not been sufficiently covered by security analyses.

Abstract In an automated machining process, monitoring the conditions of the tool is essential for deciding to replace or repair the tool without any manual intervention. Intelligent models built with sensor information and machine learning techniques are predicting the condition of the tool with good accuracy. In this study, statistical models are developed to identify the conditions of the abrasive grinding wheel using the Acoustic Emission (AE) signature acquired during the surface grinding operation. Abrasive grinding wheel conditions are identified using the abrasive wheel wear plot established by conducting experiments. The piezoelectric sensor is used to capture the AE from the grinding process, and statistical features of the abrasive wheel conditions are extracted in time and wavelet domains of the signature. Machine learning algorithms, namely, Classification and Regression Trees (CART) and Support Vector Classifiers (SVC), are used to build statistical models.

Abstract There are already a number of cybersecurity activities introduced in the development process in the automotive industry. For example, security testing of automotive components is often performed at the late stages of development. Fuzz testing is often performed as part of the security testing activity. However, since testing occurs late in the development process, it is expensive and, in some cases, may be too late to fix certain identified issues. Another challenge is that some testing requires hardware that is costly and may not be available until late in the development. We suggest fuzz testing virtual ECUs, which overcomes these challenges and allows for more efficient and effective security testing.

Abstract Management of expressway networks has been mainly focused on defect management without looking at the correlations with accidental risks. This causes unsustainability in expressway infrastructure maintenance since such defects may not be a contributing factor toward public safety. Thus it is necessary to incorporate accidental events for decision-making in infrastructure management. This study has developed a novel approach to machine learning (ML) that incorporates actual primary data from the last 10 years of single-vehicle accidents (SVA) by collisions with motorway facilities, or so-called single-vehicle collisions with fixed objects. The ML is firstly aimed at identifying the influential factors of SVA in relation to finding effective countermeasures for accidents by integrating the correlation analysis, multiple regression analysis, and ML techniques. The study reveals that wet pavement conditions have a significant effect on SVA.

Abstract Automotive cybersecurity is steadily becoming a key factor in the worldwide adoption of connected and self-driving vehicles. Following the trend set by legislation mandating standardized vehicle safety, international standards and legislation are being put into place to mandate vehicle security. Due to cultural and legal system differences, the priorities of different countries’ legislation often differ. This article seeks to explore the different approaches taken by countries in some of the major automotive markets, with a special emphasis on that of the Japanese automotive security landscape.

Abstract Connected vehicle technology consisting of Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication falls under the umbrella of V2X, or Vehicle-to-Everything, communication. This enables vehicles and infrastructure to exchange safety-related information to enable smarter, safer roads. If driver alerts are raised or automated action is taken as a result of these messages, it is critical that messages are trustworthy and reliable. To this end, the Security Credential Management System (SCMS) and Cooperative Intelligent Transportation Systems (C-ITS) Credential Management System (CCMS) have been proposed to enable authentication and authorization of V2X messages without compromising individual user privacy. This is accomplished by issuing each vehicle a large set of “pseudonyms,” unrelated to any real-world identity. During operation, the vehicle periodically switches pseudonyms, thereby changing its identity to others in the network.

Abstract A new method that allows data-enabled (empirical) models, commonly used for automotive engine calibration, to extrapolate beyond the range of training data has been developed. This method used a physics-based system-level one-dimensional model to improve interpolation and allow extrapolation for three data-based algorithms, by modifying the model input (feature) space. Neural network, regression, and k-nearest neighbor predictions of engine emissions and volumetric efficiency were greatly improved by generating 736,281 artificial feature spaces and then performing feature selection to choose feature spaces (feature selection) so that extrapolations in the original feature space were interpolations in the new feature space. A novel feature selection method was developed that used a two-stage search process to uniquely select the best feature spaces for every prediction.

Abstract One of the key factors for accurate mass burn fraction and energy conversion point calculations is the accuracy of the compression ratio. The method presented in this article suggests a workflow that can be applied to determine or correct the compression ratio estimated geometrically or measured using liquid displacement. It is derived using the observation that, in a motored engine, the heat losses are symmetrical about a certain crank angle, which allows for the derivation of an expression for the clearance volume [1]. In this article, a workflow is implemented in real time, in a current production engine indicating system. The goal is to improve measurement data quality and stability for the energy conversion points calculated during measurement procedures. Experimental and simulation data is presented to highlight the benefits and improvement that can be achieved, especially at the start of combustion.

Abstract The aim of this study is to achieve the target transient posture of a vehicle according to the user’s steering operation. The target behavior was hypothesized to be a roll mode in the diving pitch, even during steering inputs on rough surfaces, in order to improve subjective evaluation. As a result of organizing the issues of feedforward control (FF) and feedback control (FB), we hypothesized that it would be appropriate to follow the ideal posture. The model following damping control (MFDC) was newly proposed by the authors as a solution to a control algorithm based on model-following control. The feature employs skyhook control (SH), which follows the deviation between the behavior of the reference model, which generates a target behavior with no input from the road surface, and the actual behavior of the vehicle. Numerical analyses were performed to verify the followability of the target behavior and the effect of roll damping performance.

Abstract Automotive cybersecurity engineers now have the challenge of delivering Risk Assessments of their products using a method that is described in the new standard for automotive cybersecurity: International Organization for Standardization/Society of Automotive Engineers (ISO/SAE) 21434. The ISO standards are not treated in the same way as regulations that are mandated by governing bodies. However, the new United Nations (UN) Regulation No. 155 “Cyber Security and Cyber Security Management” actually drives a need to apply ISO/SAE 21434. This article investigates the practical aspects of performing such a Threat Analysis and Risk Assessment (TARA) from system modelling and asset identification to attack modelling and the consequences an attack will have.

Abstract The frequency of video-capturing collision events from surveillance systems are increasing in reconstruction analyses. The video that has been provided to the investigator may not always include a clear perspective of the relevant area of interest. For example, surveillance video of an incident may have captured a pre- or post-incident perspective that, while failing to capture the precise moment when the pedestrian was struck by a vehicle, still contains valuable information that can be used to assist in reconstructing the incident. When surveillance video is received, a quick and efficient technique to place the subject object or objects into a three-dimensional environment with a known rate of error would add value to the investigation.

Abstract Traditionally, Electronic Control Units (ECUs) in vehicles have been left unsecured. Ensuring cybersecurity in an ECU network is challenging as there is no centralized authority in the vehicle to provide security as a service. While progress has been made to address cybersecurity vulnerabilities, many of these approaches have focused on enterprise, software-centric systems and require more computational resources than typically available for onboard vehicular devices. Furthermore, vehicle networks have the additional challenge of mitigating security vulnerabilities while satisfying safety and performance constraints. This article introduces a blockchain framework to detect unauthorized modifications to vehicle ECUs. A proof of concept blockchain prototype framework is implemented on a set of microprocessors (comparable to those used by simple ECUs) as a means to assess the efficacy of using our blockchain approach to detect unauthorized updates.

Abstract In recent years, to avoid the failure of the onboard perception system, intelligent vehicle infrastructure cooperative systems have been attracting attention in the field of autonomous vehicles. Using the perception technology of roadside sensors to provide supplementary traffic information for autonomous vehicles has become an increasing trend. Several roadside perception solutions select deep learning for three-dimensional (3D) object detection. However, deep learning methods have several issues and lack reliability in practical engineering applications. To tackle this challenge, this study proposes a pedestrian detection algorithm based on roadside Light Detection And Ranging (LiDAR) by combining traditional and deep learning algorithms. To meet real-time demand, Octree with region-of-interest (ROI) selection is introduced and improved to filter the background in each frame, which improves the clustering speed.

Abstract This article describes the application of Reinforcement Learning (RL) with an embedded heuristic algorithm to a multi-objective hybrid vehicle optimization. A multi-objective optimization problem (MOP) is defined as a minimization of total energy consumption and trip time resulting from optimal control of vehicle speed over a known route. First, a computationally efficient heuristic optimization algorithm is formulated to solve the MOP for multiple traffic scenarios. Then, the off-line integration of RL is applied to the heuristic optimization algorithm process and utilized to solve the MOP. Finally, the online optimization capability of the machine learning algorithm is discussed, as well as its extension to the vehicle routing problem and the hybrid electric vehicle. The specific scenario investigated is where a generic vehicle begins a trip on a one-lane highway. The length of the highway and the number of vehicles and traffic signals on the road are generic as well.

Abstract ICVs are expected to make the transportation safer, cleaner, and more comfortable in the near future. However, the trend of connectivity has greatly increased the attack surfaces of vehicles, which makes in-vehicle networks more vulnerable to cyberattacks which then causes serious security and safety issues. In this article, we therefore systematically analyzed cyberattacks and corresponding countermeasures for in-vehicle networks of intelligent and connected vehicles (ICVs). Firstly, we analyzed the security risk of ICVs and proposed an in-vehicle network model from a hierarchical point of view. Then, we discussed possible cyberattacks at each layer of proposed network model.