Autonomous Self-driving Vehicles Research Articles

Advancements and challenges in achieving fully autonomous self-driving vehicles

This article presents a review and analysis of the prospects of achieving full-length autonomous driving, a concept that has long been a dream of humans. Although the automotive industry has made significant progress in many areas, creating fully automated vehicles (level 5) has remained a challenge. This paper examines some companies that are already racing to achieve this feat, such as Tesla, Google's Waymo, and Uber, and the challenges they face, such as ensuring safety and reliability while also dealing with complex and expensive technology. The article highlights the issues that must be addressed when discussing fully automated vehicles, such as legal and regulatory frameworks, public acceptance, and cybersecurity. The current generation of autonomous vehicles has not yet passed road tests, which raises questions about the feasibility of creating and being able to use fully automated vehicles in the near future. Despite the challenges, the article suggests that the high level of technological development implies that fully autonomous vehicles will become a reality someday. The author concludes that while we may not see fully autonomous driving cars on our roads soon, the progress being made in the field is promising, and we may expect to see these vehicles become a common sight on our roads in the future.

Equilibrium analysis of morning commuting and parking under spatial capacity allocation in the autonomous vehicle environment

This study analytically investigates the morning commuting and parking patterns of autonomous vehicles (AVs) under different spatial road capacity allocation schemes (i.e., capacity split between inbound and outbound travel directions). Given that self-driving AV might park far away from commuters’ destination, we investigate equilibrium departure/arrival and parking patterns for AVs subject to the spatial road capacity allocation. We also analyse the system optimum traffic pattern for AV morning commute under a given capacity allocation scheme. Furthermore, we examine optimal capacity allocation strategies under user equilibrium and system optimum AV traffic patterns, respectively, which aim to minimise the total system travel cost. Numerical studies are conducted to illustrate the model and analysis. The results reveal the sensitivity of different efficiency metrics with respect to AV parking supply and road capacity allocation schemes, and provide insights into the infrastructure management with future automated transport.

A 200 MHz 14 W Pulsed Optical Illuminator With Laser Driver ASIC and On-Chip DLL-Based Time Interpolator for Indirect Time-of-Flight Applications

The applications requiring three-dimensional information have been growing rapidly, including facial recognition, gesture recognition, auto-focus for 2D camera, self-driving autonomous vehicles, etc. Indirect Time-of-Flight (iToF) technique has been adopted widely for 3D depth imaging due to its high spatial resolution and excellent depth accuracy performance. In this brief, a 200 MHz pulsed optical Illuminator for iToF applications is reported. A 940 nm GaAs-based triple-junction VCSEL array is adopted and an integrated pulsed VCSEL driver ASIC in 180 nm BCD process is realized. Optical pulses with peak power up to 14 W can be transmitted by the laser illuminator with a maximum driving current of 6.5 A by the driver ASIC. The modulation frequency can be up to 200 MHz. Automatic optical power control (AOPC) is realized for stable optical power output and laser eye safety protection. An on-chip DLL-based time interpolator is proposed to eliminate the mechanical moving components used during system calibration. The measured absolute accuracy of the time interpolator is +/−50 ps with a resolution of 800 ps and tuning range of 49.6 ns. This brief effectively extends the measurement distance of the iToF system. The calibration efficiency is also improved with the proposed method.

The Use of Terrestrial and Maritime Autonomous Vehicles in Nonintrusive Object Inspection.

Traditional nonintrusive object inspection methods are complex or extremely expensive to apply in certain cases, such as inspection of enormous objects, underwater or maritime inspection, an unobtrusive inspection of a crowded place, etc. With the latest advances in robotics, autonomous self-driving vehicles could be applied for this task. The present study is devoted to a review of the existing and novel technologies and methods of using autonomous self-driving vehicles for nonintrusive object inspection. Both terrestrial and maritime self-driving vehicles, their typical construction, sets of sensors, and software algorithms used for implementing self-driving motion were analyzed. The standard types of sensors used for nonintrusive object inspection in security checks at the control points, which could be successfully implemented at self-driving vehicles, along with typical areas of implementation of such vehicles, were reviewed, analyzed, and classified.

Deep-Learning-Based Network for Lane Following in Autonomous Vehicles

The research field of autonomous self-driving vehicles has recently become increasingly popular. In addition, motion-planning technology is essential for autonomous vehicles because it mitigates the prevailing on-road obstacles. Herein, a deep-learning-network-based architecture that was integrated with VGG16 and the gated recurrent unit (GRU) was applied for lane-following on roads. The normalized input image was fed to the three-layer VGG16 output layer as a pattern and the GRU output layer as the last layer. Next, the processed data were fed to the two fully connected layers, with a dropout layer added in between each layer. Afterward, to evaluate the deep-learning-network-based model, the steering angle and speed from the control task were predicted as output parameters. Experiments were conducted using the a dataset from the Udacity simulator and a real dataset. The results show that the proposed framework remarkably predicted steering angles in different directions. Furthermore, the proposed approach achieved higher mean square errors of 0.0230 and 0.0936 and and inference times of 3–4 and 3 ms. We also implemented our proposed framework on the NVIDIA Jetson embedded platform (Jetson Nano 4 GB) and compared it with the GPU’s computational time. The results revealed that the embedded system took 45–46 s to execute a single epoch in order to predict the steering angle. The results show that the proposed framework generates fruitful and accurate motion planning for lane-following in autonomous driving.

Managing mixed traffic with autonomous vehicles – A day-to-day routing allocation scheme

In presence of the emerging technology of automated vehicles, it is anticipated that the future traffic system would be comprised of mixed traffic with both self-driving autonomous vehicles (AVs) and human-driven conventional vehicles. It is imperative to propose new traffic management measures to manage the future traffic system, as complement of the existing ones such as road pricing schemes. This study seeks to take advantage of the controllable property of AVs’ routing choices to develop a day-to-day routing allocation scheme for a certain number of autonomous vehicles so as to drive the mixed traffic system into a desired traffic state. Specifically, we assume that all travelers are bounded rational and AV users are willing to accept route allocation with route travel cost not exceeding HV users’ indifference band. Therefore, the best-case bounded rationality user equilibrium (BRUE) flow pattern is in principle the most desirable traffic state out of all the BRUE solutions. This study proposes a day-to-day AVs’ routing allocation scheme by which the traffic system would be directed to evolve towards the desired best-case BRUE. The day-to-day traffic dynamics of HVs are proposed to follow a general framework, which can be further proved to be the BRUE rational behavior adjustment process (BRUE-RBAP). The condition for convergence is investigated under general assumptions. Numerical examples are provided to demonstrate the effectiveness of this traffic management scheme.

Autonomous self-driving vehicles - Design of professional laboratory exercises in the field of automotive mechatronics

Self-driving cars are gradually making their way into road traffic and represent the main component of the new form of mobility. Major companies such as Tesla, Google, and Uber are researching the continuous improvement of self-driving vehicles and their reliability. Therefore, it is of great interest for trained professionals to deal with and understand the principles and requirements for autonomous driving. This paper aims to describe the new concept of a Bachelor / Master level university course for automotive technology students to address new mobility and self-driving cars. For the practice-oriented course, hardware in a low budget range (US $80) was used, which nevertheless has all the necessary sensors and requirements for a comprehensive practical introduction to the topic of self-driving automotive technology. The modular structure of the course contains lectures and exercises on the following topics: The first Exercise deals with the construction and modification of the Car-Kit, followed by the setup of the used Raspberry Pi. Since the car kit and Raspberry Pi are ready to use, the third exercise will steer the car remotely. The autonomous lane lectures and exercises follow this navigation with color spaces and masking, the Canny edge detection, Hough transform, steering, and stabilization.

Introduction: the secret lives of microbial mobile genetic elements.

Introduction: the secret lives of microbial mobile genetic elements.

Exploring the Potential of Using Privately-Owned, Self-Driving Autonomous Vehicles for Evacuation Assistance

The potential use of privately-owned autonomous vehicles (AVs) for the evacuation of carless households threatened by hurricanes is underexplored. Based on 518 original survey responses from South Carolina (SC) residents, an ordered logistic model was developed to determine the willingness of individuals to temporarily share their AVs for evacuation without their presence. The model results indicated that respondents who (a) were unemployed, (b) had experience giving disaster relief assistance, (c) took regular religious trips and were more comfortable with AVs (d) delivering packages and (e) being purchased and shared for income in the next five years were more willing to share for evacuation. Respondents who (a) were aged 65 or older, (b) had income below $ 15,000 per year, and (c) had less than two social media accounts were less willing to share. The model was applied to a state-wide synthetic population to simulate a disaster scenario in SC under different AV market penetration ( p ) scenarios to determine the potential use of AVs for evacuation assistance. Monte Carlo simulation results indicated that the percentage of households that can be evacuated increased linearly with respect to p , by 5.5 % for every 1 % increase in p until p was nearly 20 % . When p was 30 % or higher, the number of shared AVs was sufficient to evacuate all households in need. Therefore, in SC, if privately-owned AVs are widely available, they could serve as a viable alternative or be used to supplement the traditional evacuation programs that rely on buses.

Robust super-resolution depth imaging via a multi-feature fusion deep network.

The number of applications that use depth imaging is increasing rapidly, e.g. self-driving autonomous vehicles and auto-focus assist on smartphone cameras. Light detection and ranging (LIDAR) via single-photon sensitive detector (SPAD) arrays is an emerging technology that enables the acquisition of depth images at high frame rates. However, the spatial resolution of this technology is typically low in comparison to the intensity images recorded by conventional cameras. To increase the native resolution of depth images from a SPAD camera, we develop a deep network built to take advantage of the multiple features that can be extracted from a camera's histogram data. The network is designed for a SPAD camera operating in a dual-mode such that it captures alternate low resolution depth and high resolution intensity images at high frame rates, thus the system does not require any additional sensor to provide intensity images. The network then uses the intensity images and multiple features extracted from down-sampled histograms to guide the up-sampling of the depth. Our network provides significant image resolution enhancement and image denoising across a wide range of signal-to-noise ratios and photon levels. Additionally, we show that the network can be applied to other data types of SPAD data, demonstrating the generality of the algorithm.

Customer Acceptance of Autonomous Vehicles in Travel and Tourism

Since the COVID-19 pandemic has significantly increased the use of personal vehicles in travel, adoption of self-driving autonomous vehicles can radically transform the travel industry. Thus, this study develops and tests a conceptual autonomous vehicle acceptance model that identifies hedonic motivation, trust in autonomous vehicles and social influence as critical determinants of performance expectancy, perceived risk and emotions, which determine travelers’ intentions to use autonomous vehicles (AVs) utilizing the Cognitive Appraisal Theory and the Artificially Intelligent Device Use Acceptance model as conceptual frameworks. Findings indicate that trust is the most powerful determinant of performance expectancy and essential to decrease risk perceptions. Furthermore, performance expectancy and hedonic motivation are critical determinants of travelers’ positive emotions, which in turn determines the acceptance of AVs. Contribution to theoretical knowledge and implications for practice are provided, and limitations and recommendations for future studies are discussed.

Are Self-Driving Vehicles Ready to Launch? An Insight into Steering Control in Autonomous Self-Driving Vehicles

In the last couple of years, academia-industry collaboration has demonstrated rapid advancements in the development of self-driving vehicles. Since it is evident that self-driving vehicles are going to reshape the traditional transportation systems in near future through enhancement in safe and smart mobility, motion control in self-driving vehicles while performing driving tasks in a dynamic road environment is still a challenging task. In this regard, we present a comprehensive study considering the evolution of steering control methods for self-driving vehicles. Initially, we discussed an insight into the traditional steering systems of the vehicles. To the best of our knowledge, currently, there is no taxonomy available, which elaborates steering control methods for self-driving vehicles. In this paper, we present a novel taxonomy including different steering control methods which are categorized into deterministic and heuristic steering control methods. Concurrently, the abovementioned techniques are critically reviewed elaborating their strengths and limitations. Based on the analysis, key challenges/research gaps in existing steering control methods along with the possible solutions have been briefly discussed to improve the effectiveness of the steering system of self-driving vehicles.

Coordination of Autonomous Vehicles

In the near future, our streets will be populated by myriads of autonomous self-driving vehicles to serve our diverse mobility needs. This will raise the need to coordinate their movements in order to properly handle both access to shared resources (e.g., intersections and parking slots) and the execution of mobility tasks (e.g., platooning and ramp merging). The aim of this article is to provide a global view of the coordination issues and the related solutions in the field of autonomous vehicles. To this end, we firstly introduce the general problems associated with coordination of autonomous vehicles by identifying and framing the key classes of coordination problems. Then, we overview the different approaches that can be adopted to deal with such problems by classifying them in terms of the degree of autonomy in decision making that is left to autonomous vehicles during the coordination process. Finally, we overview some further research challenges to address before autonomous coordinated vehicles can safely hit our streets.

Conceptualisation of Human-on-the-Loop Haptic Teleoperation With Fully Autonomous Self-Driving Vehicles in the Urban Environment

The automotive industry aims to deploy commercial level-5 fully autonomous self-driving vehicles (FA-SDVs) in a diverse range of benefit-driven concepts on city roads in the years to come. In all future visions of operating networks of FA-SDVs, humans are expected to intervene with some kind of remote supervisory role. Recent advances in cyber-physical systems (CPS) within the concept of Internet of Everything (IoE) using tactile internet (TI) teleport us to teleoperate remote objects within the cyber-world. Human-on-the-loop (HOTL) haptic teleoperation with an extension of human control and sensing capability by coupling with artificial sensors and actuators with an increased sense of real-time driving in the remote vehicle can help overcome the challenging tasks when the new driver - artificial intelligence (AI) agent - encounters an unorthodox situation that can't be addressed by the autonomous capabilities. This paper analyses HOTL real-time haptic delay-sensitive teleoperation with FA-SDVs, in the aspects of human-vehicle teamwork by establishing two similar remote parallel worlds --- real-world vehicle time-varying environment and cyber-world emulation of this environment, i.e., digital twins (DTs) --- in which a human telesupervisor (HTS), as a biological agent, can be immersed within a reasonable timescale with no cybersickness enabling omnipresence and a bidirectional flow of energy and information. The experiments conducted as a proof of concept of HOTL haptic teleoperation shows promising results and the potential of benefiting from the proposed framework.

Consumption Patterns and the Advent of Automated Guided Vehicles, and the Trends for Automated Guided Vehicles

In this paper we review how automation and robots are a key player in contributing to the increased need for profit by providing decrease production costs, reducing delivery time, improving quality, and increasing customer satisfaction. In particular, we focus on logistics robots, namely, automated guided vehicles (AGVs) and automated mobile robots (AMRs), generically addressing their applications, state of the art, and trends. We further identify technological trends and implementation challenges. Finally, we discuss how they can influence future implementations of the Factory of the Future and help attaining the goal of implementing all lights out factory. The most typical applications of AGVs are transportation (raw materials from warehouse to production line and finished products back to warehouse), storage (holding or buffering material over a period of time), distribution (fulfillment, cross-dock facility, bulk break, and package handling) and assembly (reduce the line-side storage requirements, enabling just-in-time). AGVs are a complex systems’ engineering product based on integrating many different technologies. We address commonly used technologies for navigation, positioning, and management, among others, and present novel technologies and development trends. The 4 Industrial Revolutions and the ever-increasing consumption demands have been putting pressure on both production plants and logistics chains. With e-commerce surging and tele-work being redefined to new scales, the efficiency and speed of manufacturing and delivery are paramount. Although AGVs are already used in many industries and applications, self-driving and autonomous vehicles still have a bright, untapped future ahead, but with many technical and regulatory challenges to tackle. The market will grow even faster as more technologies are incorporated at lower prices and as it becomes easier to deploy and commission AGVs. On the other hand, the fusion and intelligent cooperation between standard automated vehicles and autonomous self-driving vehicles will open new market segments of increasingly significant importance. We have thus concluded that although AGVs have been gaining traction in the past decade, the democratization of implementing AGVs is far from being around the corner. The integration of AGVs in different manufacturing and logistics processes will skyrocket when installing and deploying an AGV is effortless, and it is likely that speed of adoption of AGVs will benefit and leverage on the increasingly more advanced and more readily available software for autonomous self-driving vehicles.

Secure decentralized peer-to-peer training of deep neural networks based on distributed ledger technology

The accuracy and performance of deep neural network models become important issues as the applications of deep learning increase. For example, the navigation system of autonomous self-driving vehicles requires very accurate deep learning models. If a self-driving car fails to detect a pedestrian in bad weather, the result can be devastating. If we can increase the model accuracy by increasing the training data, the probability of avoiding such scenarios increases significantly. However, the problem of privacy for consumers and lack of enthusiasm for sharing their personal data, e.g., the recordings of their self-driving car, is an obstacle for using this valuable data. In Blockchain technology, many entities which cannot trust each other in normal conditions can join together to achieve a mutual goal. In this paper, a secure decentralized peer-to-peer framework for training the deep neural network models based on the distributed ledger technology in Blockchain ecosystem is proposed. The proposed framework anonymizes the identity of data providers and therefore can be used as an incentive for consumers to share their private data for training deep learning models. The proposed framework uses the Stellar Blockchain infrastructure for secure decentralized training of the deep models. A deep learning coin is proposed for Blockchain compensation.

Smart City: Traffic Management System Using Smart Sensor Network

by 2050, 70% of the world population is estimated to migrate toward the city, covering just 2% of the earth’s surface. This leads to many issues, and traffic congestion is one among them. To continue to serve and improve the life of the growing population, it is mandatory to develop an advanced medium to avoid risks which are likely to occur as a result of overcrowded traffic. Our motive is to develop an autonomous vehicle to free human drivers and thus increasing their safety. The Self-driving autonomous vehicles have intensely become one of the great discovery in the field of technology. Different technologies like deep learning, Artificial Intelligence (AI), etc. are merged with each other and with the smart sensors (SS) to develop this self-driven autonomous vehicle. Computer Vision and Deep Learning techniques are applied to build an automotive related algorithm. This project uses Computer Vision Techniques to identify lane lines on a road and also able to identify 40+ different traffic signals. In this project, we will explore the smart city (SMC) concept and propose a strategy development model that will mitigate the traffic concern issues by implementing traffic management system (TMS) using SS network in a SMC context.

A Comparative Study of Different Objectives Functions for the Minimal Fuel Drive Cycle Optimization in Autonomous Vehicles

With the advent of self-driving autonomous vehicles, vehicle controllers are free to drive their own velocities. This feature can be exploited to drive an optimal velocity trajectory that minimizes fuel consumption. Two typical approaches to drive cycle optimization are velocity smoothing and tractive energy minimization. The former reduces accelerations and decelerations, and hence, it does not require information of vehicle parameters and resistance forces. On the other hand, the latter reduces tractive energy demand at the wheels of a vehicle. In this work, utilizing an experimentally validated full vehicle simulation software, we show that for conventional gasoline vehicles the lower energy velocity trajectory can consume as much fuel as the velocity smoothing case. This implies that the easily implementable, vehicle agnostic velocity smoothing optimization can be used for velocity optimization rather than the nonlinear tractive energy minimization, which results in a pulse-and-glide trajectory.

Assess the impacts of different autonomous trucks’ lateral control modes on asphalt pavement performance

With ongoing massive investment in the field of autonomous vehicles, fully autonomous self-driving vehicles are set to become a reality soon. Despite the benefits in traffic safety and economy, the potential effects of autonomous vehicles, especially autonomous trucks, on transportation infrastructure remain to be determined. In this study, a finite element model was developed based on a typical flexible pavement structure. The effects of autonomous trucks’ lateral distribution within the lane with respect to rutting depth and fatigue damage were estimated by finite element analysis under certain environmental condition. Characteristics of the pavement structure and environmental condition have been collected and presented in this paper. Four possible lateral control modes on managing autonomous trucks’ lateral distribution were proposed, listed as zero-wander mode, uniform mode, double peak Gaussian mode, two-section uniform mode. Considering the difference between autonomous trucks and human-driven trucks, the ratio of autonomous trucks was also evaluated. Based on the simulation results, it has been confirmed that, if controlled appropriately, autonomous trucks could be highly beneficial to asphalt pavements for wider using of pavement. Conversely, if without appropriate lateral control, the effect of autonomous trucks on the typical flexible pavement life could be negative, owing to repeated single point load caused by lane centering and keeping. For example, the uniform mode, double peak Gaussian mode and two-section uniform mode all throw positive effect on the pavement. On the contrary, under the zero wander mode, the time when the rutting depth of the proposed pavement reaches 15 mm (defined as maintenance year) may advance by 1.56 years and fatigue damage at the bottom of asphalt layer increases by 146%. Compared with the normal situation nowadays, the two-section uniform distribution of autonomous trucks, the best performing control mode, delays the maintenance year by 2.3 years. From the aspect of fatigue, a much uniform distribution curve of fatigue within the lane can also be found under this mode. When all trucks’ lateral positions were in control, the two-section uniform mode could reduce the fatigue damage by up to 35% under repeated standard axle load on the flexible pavement. Finally, a framework of dealing with the lateral control was proposed as a reference for different conditions, such as asphalt pavement structures and materials.

Cloud Computing-Based Analyses to Predict Vehicle Driving Shockwave for Active Safe Driving in Intelligent Transportation System

In the cloud computing era, the analyses of various types of big data gathered from the Internet of Vehicles, the Internet of Things, and smart sensors/devices achieve convenient services. Some famous applications include vehicle traffic loading on roads (from Google MAP), video sharing for fans group (from Facebook or Youtube), and so on. However, to achieve real-time active safe driving (RT-ASD) under unstable driving in high-threat areas on roads becomes an open issue in cloud computing-based intelligent transportation system. This paper proposes a predictive backward shockwave analysis approach (PSA) to achieve the RT-ASD based on the analyses of macroscopic traffic shockwave ( PSA_MA ) and microscopic car-following ( PSA_mi ). The PSA contributes in several aspects to active safe driving: 1) predicting and analyzing high threat of backward shockwaves from the gathered big data of the driving state information vehicles; 2) informing the analyzed threat messages to the vehicles in high-threat areas via the 3-Tier hierarchical cloud computing mechanism; 3) reducing the driving threat certainly; and 4) PSA_MA and PSA_mi can be applied for achieving active safe driving in autonomous self-driving vehicles and human-driving vehicles. The numerical results show that the PSA outperforms in approaches to relative error rate prediction, the accuracy of the backward shockwave determination, average vehicle velocity, average travel time, number of goodput vehicles, time-to-collision, and distance-to-collision.