Reliability Of Autonomous Vehicles Research Articles

Ontology of autonomous driving based on the SAE J3016 standard

Autonomous driving is a recently developed area in which technology seems to be ahead of its understanding within society. That causes some fears concerning the reliability of autonomous vehicles and controversies over liability in case of accidents. Specifying levels of driving autonomy within the SAE-J3016 standard is widely recognized as a significant step towards comprehending the essence of the achievements. However, the standard provides even more valuable insights into the process of driving automation. In the paper, we develop the ideas using the methods of formal ontology that allow us to make the conceptual system more precise and formalize it. To increase inseparability, we ground our system on a top-level BFO ontology. We present a formal account of several areas covered by the SAE-J3016 standard, including motor vehicles and their systems, driving tasks and subtasks, roles of persons in road communication, and autonomy levels.

The Road Ahead: A Survey of Autonomous Vehicle Challenges and Opportunities

This comprehensive study delves into the rapidly advancing field of autonomous vehicles (AVs). We explore the core technologies that enable AVs to navigate, perceive their surroundings, make informed decisions, and control their movements. The intricate interplay between sensor fusion, localization and mapping, planning algorithms, and control systems is examined. A critical focus of this research is on the ethical and societal implications of AV deployment. We analyze the crucial considerations surrounding safety, security, legal frameworks, and ethical guidelines that must be addressed to ensure responsible adoption of AVs. The potential social and economic impacts of widespread AV adoption are examined, including implications for transportation infrastructure, job markets, and urban planning .Furthermore, this study explores the exciting future directions of AV technology. We discuss emerging trends and innovations, such as the integration of artificial intelligence and machine learning, advancements in sensor technology, and the development of advanced communication networks. These developments hold the promise of further enhancing the capabilities and reliability of AVs, paving the way for a future where autonomous transportation becomes a ubiquitous part of our daily lives. Keywords and Phrases: Autonomous Vehicles, Sensor Fusion, Localization and Mapping, Planning Algorithms, Control Systems, Safety, Security, Legal Frameworks, Ethical Considerations, Social Impact, Economic Impact, Future Directions.

Enhanced Object Detection in Autonomous Vehicles through LiDAR—Camera Sensor Fusion

To realize accurate environment perception, which is the technological key to enabling autonomous vehicles to interact with their external environments, it is primarily necessary to solve the issues of object detection and tracking in the vehicle-movement process. Multi-sensor fusion has become an essential process in efforts to overcome the shortcomings of individual sensor types and improve the efficiency and reliability of autonomous vehicles. This paper puts forward moving object detection and tracking methods based on LiDAR—camera fusion. Operating based on the calibration of the camera and LiDAR technology, this paper uses YOLO and PointPillars network models to perform object detection based on image and point cloud data. Then, a target box intersection-over-union (IoU) matching strategy, based on center-point distance probability and the improved Dempster–Shafer (D–S) theory, is used to perform class confidence fusion to obtain the final fusion detection result. In the process of moving object tracking, the DeepSORT algorithm is improved to address the issue of identity switching resulting from dynamic objects re-emerging after occlusion. An unscented Kalman filter is utilized to accurately predict the motion state of nonlinear objects, and object motion information is added to the IoU matching module to improve the matching accuracy in the data association process. Through self-collected data verification, the performances of fusion detection and tracking are judged to be significantly better than those of a single sensor. The evaluation indexes of the improved DeepSORT algorithm are 66% for MOTA and 79% for MOTP, which are, respectively, 10% and 5% higher than those of the original DeepSORT algorithm. The improved DeepSORT algorithm effectively solves the problem of tracking instability caused by the occlusion of moving objects.

GPT-4V Takes the Wheel: Promises and Challenges for Pedestrian Behavior Prediction

Predicting pedestrian behavior is the key to ensure safety and reliability of autonomous vehicles. While deep learning methods have been promising by learning from annotated video frame sequences, they often fail to fully grasp the dynamic interactions between pedestrians and traffic, crucial for accurate predictions. These models also lack nuanced common sense reasoning. Moreover, the manual annotation of datasets for these models is expensive and challenging to adapt to new situations. The advent of Vision Language Models (VLMs) introduces promising alternatives to these issues, thanks to their advanced visual and causal reasoning skills. To our knowledge, this research is the first to conduct both quantitative and qualitative evaluations of VLMs in the context of pedestrian behavior prediction for autonomous driving. We evaluate GPT-4V(ision) on publicly available pedestrian datasets: JAAD and WiDEVIEW. Our quantitative analysis focuses on GPT-4V's ability to predict pedestrian behavior in current and future frames. The model achieves a 57% accuracy in a zero-shot manner, which, while impressive, is still behind the state-of-the-art domain-specific models (70%) in predicting pedestrian crossing actions. Qualitatively, GPT-4V shows an impressive ability to process and interpret complex traffic scenarios, differentiate between various pedestrian behaviors, and detect and analyze groups. However, it faces challenges, such as difficulty in detecting smaller pedestrians and assessing the relative motion between pedestrians and the ego vehicle.

An in-depth evaluation of deep learning-enabled adaptive approaches for detecting obstacles using sensor-fused data in autonomous vehicles

This paper delivers an exhaustive analysis of the fusion of multi-sensor technologies, including traditional sensors such as cameras, Light Detection and Ranging(LiDAR), Radio Detection and Ranging(RADAR), and ultrasonic sensors, with Artificial Intelligence(AI) powered methodologies in obstacle detection for Autonomous Vehicles(AVs). With the growing momentum in AVs adoption, a heightened need exists for versatile and resilient obstacle detection systems. Our research delves into study of literatures, where proposed approaches assimilate data from this diverse sensor suite, integrated through Deep Learning(DL) techniques, to refine AV performance. Recent advancements and prevailing challenges within the domain are thoroughly examined, with particular focus on the integration of sensor fusion techniques, the facilitation of real-time processing via edge and fog computing, and the implementation of advanced artificial intelligence architectures, including Convolutional Neural Networks(CNNs), Recurrent Neural Networks(RNNs), and Generative Adversarial Networks(GANs), to enhance data interpretation efficacy. In conclusion, the paper underscores the critical contribution of multi-sensor arrays and deep learning in enhancing the safety and reliability of autonomous vehicles, offering significant perspectives for future research and technological progress.

Intrusion Detection System for Autonomous Vehicles Using Non-Tree Based Machine Learning Algorithms

This research work highlights significant achievements in the domain of intrusion detection systems (IDSs) for autonomous vehicles, which are crucial in enhancing their safety, reliability, and cybersecurity. This study introduces an approach that leverages non-tree-based machine learning algorithms, such as K-nearest neighbors and ensemble learning, to develop an IDS tailored for autonomous vehicles. These algorithms were employed because of their ability to process complex and large datasets with less likeliness for overfitting, their scalability, and their ability to adapt to changing conditions in real time. These algorithms effectively handle imbalanced data, enhancing the detection accuracy of both normal and intrusive instances. The IDS’s performance was validated through the utilization of three real-world datasets, CAN intrusion, CICIDS2017, and NSL-KDD, where the proposed non-tree-based IDS (NTB-MTH-IDS) was measured with the standard measurement metrics: accuracy, precision, F1-score, and recall, including specificity and sensitivity. Notably, the results indicate that K-nearest neighbors and stacking, as part of NTB-MTH-IDS, has an accuracy of 99.00%, 98.57%, and 97.57%, and F1-scores of 99.00%, 98.79%, and 97.54% in the CICIDS2017, NSL-KDD, and CAN datasets, respectively. The results of this research can lead to establishing a robust intrusion detection framework, thereby ensuring the safety and reliability of autonomous vehicles. Through this achievement, road users, passengers, and pedestrians are safeguarded against the consequences of potential cyber threats.

Authority Transfer According to a Driver Intervention Intention Considering Coexistence of Communication Delay

Recently, interest and research on autonomous driving technology have been actively conducted. However, proving the safety of autonomous vehicles and commercializing autonomous vehicles remain key challenges. According to a report released by the California Department of Motor Vehicles on self-driving, it is still hard to say that self-driving technology is highly reliable. Until fully autonomous driving is realized, authority transfer to humans is necessary to ensure the safety of autonomous driving. Several technologies, such as teleoperation and haptic-based approaches, are being developed based on human-machine interaction systems. This study deals with teleoperation and presents a way to switch control from autonomous vehicles to remote drivers. However, there are many studies on how to do teleoperation, but not many studies deal with communication delays that occur when switching control. Communication delays inevitably occur when switching control, and potential risks and accidents of the magnitude of the delay cannot be ignored. This study examines compensation for communication latency during remote control attempts and checks the acceptable level of latency for enabling remote operations. In addition, supplemented the safety and reliability of autonomous vehicles through research that reduces the size of communication delays when attempting teleoperation. It is expected to prevent human and material damage in the actual accident situation.

Reliability Analysis of the Functional Capabilities of an Autonomous Vehicle

The reliability of autonomous vehicles (AVs) is a research domain of high interest, covering a diverse pool of researchers, captains of smart auto industries, government agencies, and technology enthusiasts. The reliability of AVs is not extensively explored in the literature, despite the apprehension due to fatal accidents recorded in the past. Despite being in existence for over a decade, AVs have yet to reach a certified commercial-level deployment. Due to the complexity that comes with the self-operation of an AV, the issue of trustworthiness, which signifies reliability, becomes inevitable. The identification, analysis, and categorization of functional elements using systems engineering conceptual design principles and the linkage of these to the road traffic rules were conducted to address this. Also, the evaluation of the reliability of AVs using various developed vehicles from selected industries was addressed by integrating the traffic rules. The analysis of reliability was carried out using life-to-failure data premised on the probability plotting approach. It was found that there is a 99.94% chance that an autonomous vehicle will fail at least one of the traffic rules within 20 minutes of driving. Furthermore, the hazard rate of AVs was found to be on the rise, meaning a high indication of accidents.

Research on global path planning of unmanned vehicles based on improved ant colony algorithm in the complex road environment

When planning the path of a non-urbanized road, the default ant colony optimization (ACO) algorithm does not consider complex road state function such as uneven surface, road attachment coefficient, and vehicle turning angle limit. Based on the actual situation of roads and vehicles, a pavement state function that considers uneven areas such as road bumps and pavement attachment is proposed to improve the description of path length. Then, a heuristic function based on the A* algorithm and an improved mechanism for the initialization of pheromone distribution is proposed, which changes the blindness of ant colony search, accelerates the convergence of the ACO, and improves the search efficiency. The global search capability of the algorithm is enhanced by improving the path selection strategy and path transition probability function. The pheromone updating method is improved by using the MAX-MIN Ant System, which increases the algorithm diversity and avoids local optima. Further, using the pruning algorithm to reduce the number of paths significantly increases the convergence speed. Simulation results show that the improved ACO algorithm has better convergence speed and global search ability. Combining road state processing with vehicle corner processing can effectively improve the safety, adaptability, and reliability of autonomous vehicles. And the global optimal path planning of unmanned vehicles on complex roads can also be realized.

Road Safety Analysis of Autonomous Vehicles

For the widespread use of Autonomous Vehicles (AVs), a huge number of challenges must be solved by vehicle manufacturers, in contrast they do have significant potential to increase road safety in both passenger and freight transport. In addition to reducing road traffic accidents and traffic jams, AVs also offer a major opportunity to reduce pollutant emissions and CO2 emissions from environmental point of view. In order to implement accident-free traffic, also called Vision Zero, it is essential to examine the safety and reliability of AVs. This article analyzes road traffic accident data and the potential safety benefits of AVs. Furthermore, the paper also sets the safety of the conventional vehicles against AVs and examines the type, location, causes, and dynamics of the accidents. The article also provides an overview over the current development trends and challenges, such as the risk of cyber-attacks, the necessary improvements in sensing technologies, and the not insignificant moral issue of AVs.

DeepSuite: A Test Suite Optimizer for Autonomous Vehicles

Deep learning (DL) brings autonomous vehicles (AVs) close to reality. However, the witness of many safety issues has raised a big concern about the reliability of AVs. To solve this problem, much research has been done to test deep learning-driven AVs. Generally, once a test input is produced, a developer needs to manually check its expected output. However, there often exists massive unlabeled test data (e.g., raw context traces in the real world). It is impractical to manually label all test inputs. Despite some works on automatic generation of test oracles, they are either task-specific or constrained to synthetic inputs. In this paper, we present a general and extensible framework, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DeepSuite</i> , to mitigate the manual effort of generating test oracles. The intuition behind is that not all test inputs are equally worth labelling. With limited testing budget, it is desirable to label a test suite with high diversity and a reasonable size. Due to the large search space, to optimize such test suites is of great challenge. To address it, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DeepSuite</i> employs a three-phase optimization method (i.e., selection, crossover, and mutation) to iteratively select representative but non-redundant test suites. Such conflicting profit/cost objectives are attained through a genetic algorithm with a well-defined multi-objective fitness function. In the experiments, we first show that the diversity of tests can be revealed by test criteria. Then, experiments on three widely-used datasets demonstrated the effectiveness of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DeepSuite</i> in generating test suites with competitive testing coverage and 68.42% smaller size, which greatly improves the data collection efficiency of testing DL-driven autonomous vehicles.

H-V2X Mode 4 Adaptive Semipersistent Scheduling Control for Cooperative Internet of Vehicles

Due to the rapidly growing number of autonomous vehicles, establishing an intelligent transportation system (ITS) supported by the Internet-of-Vehicles (IoV) technology is important. The need for IoV technology is emphasized even more because recent accidents relating to autonomous vehicles show that there are limitations in the reliability of autonomous vehicles only using vehicle onboard sensors. In addition, a highly reliable support system for many vehicular applications is in strong demand, which is why there is a need for advanced vehicle-to-everything (V2X) technology. The two major types of V2X technology are dedicated short-range communication (DSRC) and cellular-V2X (C-V2X). Many vehicles are already equipped with DSRC; however, the use of C-V2X is highly supported by many mobile communication companies and the 5GAA, which is why C-V2X is expected to be very widely used. In this article, a hybrid V2X (H-V2X) scheme is proposed, which adaptively controls C-V2X to satisfy the key-performance indicator (KPI) requirements as much as possible, and uses DSRC transmission when the C-V2X required performance cannot be reliably supported. The performance analysis graphs show that the gain in performance of the proposed H-V2X scheme can provide an improved performance compared to other V2X schemes, which makes it more suitable to support reliable autonomous vehicles.

Time-to-contact control: improving safety and reliability of autonomous vehicles

Under traditional car-following control, i.e., human drivers' behaviour, the stability condition of traffic system is not satisfied in general. For safety and reliability of autonomous vehicles, additional danger warning system must be used in the adaptive cruise control system to prevent inevitable potential collisions. One reasonable quantity of evaluating potential collisions is time to contact (TTC): how soon will potential collision occur? In this paper, we provide TTC feedback control to improve safety and reliability of autonomous vehicles, and show the effectiveness of TTC feedback. TTC can be estimated by machine vision techniques with single uncelebrated camera (i.e., passive sensors). We provide detailed mathematical analysis and algorithmic implementation. The machine vision-based TTC algorithm is pretty fast such that the whole system can be implemented on Android smartphones running in real-time. Moreover, it is not trial to estimate relative velocity by differentiating the measured distance between cars with respect to time, because inevitable measurement noise in the distance measurements will be amplified by the derivative operation. The time-to-contact-based algorithm provides an alternative approach to estimating the relative velocity, which can also be fused with measurements from other active sensors, if desired.

Time-to-contact control: improving safety and reliability of autonomous vehicles

Time-to-contact control: improving safety and reliability of autonomous vehicles

Evolution of a Reliable and Extensible High-Level Control System for an Autonomous Car

The reliability of autonomous vehicles is heavily dependent on their software frameworks, which are required to interface and process data from many different sensors on board the vehicle, perform navigational processes such as path planning and lane keeping, take action to ensure safety and display data to an operator in a useful fashion. These sensors can include a combination of cameras, LiDARs, GPS, IMU, and odometric sensors to achieve positioning and localisation for the vehicle and nearby objects in their environment and can be challenging to integrate. In this paper, we present a hybridised software framework that combines sensor and navigational processing for autonomous driving. Our framework utilises a modular approach for interfacing and safety functionality, whilst navigation and sensor interfaces are implemented as nodes in the robot operating system. Our testing results verify the suitability of our framework by integration with a hardware-in-the-loop simulation system and for fully autonomous field driving.

IDH-CAN: A Hardware-Based ID Hopping CAN Mechanism With Enhanced Security for Automotive Real-Time Applications

Cybersecurity is increasingly important for the safety and reliability of autonomous vehicles. The controller area network (CAN) is the most widely used in-vehicle network for automotive safety-critical applications. Enhancing the cybersecurity ability of CAN while considering the real-time, schedulability, and cost constraints becomes an urgent issue. To address this problem, a real-time, and schedulability analysis-guaranteed security mechanism [identification hopping CAN (IDH-CAN)] is proposed in this paper, which aims to improve the security performance of CAN under the constraints of automotive real-time applications. In order to support the operation of the IDH-CAN mechanism, an IDH-CAN controller is also designed and implemented on a field-programmable gate array, which can work as a hardware firewall in the data link layer of CAN to isolate cyberattacks from the physical layer. Meanwhile, to maximize the information entropy of the CAN message ID on the physical layer, the ID hopping table generation and optimization algorithms for IDH-CAN are also proposed. Then, information security evaluation experiments based on information entropy comparison are deployed. The simulation and practical evaluations demonstrate the effectiveness of the proposed mechanism in defending reverse engineering, targeted denial of service, and replay attacks without violating real-time and schedulability constraints.

Influence of connected and autonomous vehicles on traffic flow stability and throughput

The introduction of connected and autonomous vehicles will bring changes to the highway driving environment. Connected vehicle technology provides real-time information about the surrounding traffic condition and the traffic management center’s decisions. Such information is expected to improve drivers’ efficiency, response, and comfort while enhancing safety and mobility. Connected vehicle technology can also further increase efficiency and reliability of autonomous vehicles, though these vehicles could be operated solely with their on-board sensors, without communication. While several studies have examined the possible effects of connected and autonomous vehicles on the driving environment, most of the modeling approaches in the literature do not distinguish between connectivity and automation, leaving many questions unanswered regarding the implications of different contemplated deployment scenarios. There is need for a comprehensive acceleration framework that distinguishes between these two technologies while modeling the new connected environment. This study presents a framework that utilizes different models with technology-appropriate assumptions to simulate different vehicle types with distinct communication capabilities. The stability analysis of the resulting traffic stream behavior using this framework is presented for different market penetration rates of connected and autonomous vehicles. The analysis reveals that connected and autonomous vehicles can improve string stability. Moreover, automation is found to be more effective in preventing shockwave formation and propagation under the model’s assumptions. In addition to stability, the effects of these technologies on throughput are explored, suggesting substantial potential throughput increases under certain penetration scenarios.