Development Of Autonomous Vehicles Research Articles

A stochastic two-dimensional intelligent driver car-following model with vehicular dynamics

The law of vehicle movement has long been studied under the umbrella of microscopic traffic flow models, especially the car-following (CF) models. These models of the movement of vehicles serve as the backbone of traffic flow analysis, simulation, autonomous vehicle development, etc. Two-dimensional (2D) vehicular movement is basically stochastic and is the result of interactions between a driver’s behavior and a vehicle’s characteristics. Current microscopic models either neglect 2D noise, or overlook vehicle dynamics. The modeling capabilities, thus, are limited, so that stochastic lateral movement cannot be reproduced. The present research extends an intelligent driver model (IDM) by explicitly considering both vehicle dynamics and 2D noises to formulate a stochastic 2D IDM model, with vehicle dynamics based on the stochastic differential equation (SDE) theory. Control inputs from the vehicle include the steer rate and longitudinal acceleration, both of which are developed based on an idea from a traditional intelligent driver model. The stochastic stability condition is analyzed on the basis of Lyapunov theory. Numerical analysis is used to assess the two cases: (i) when a vehicle accelerates from a standstill and (ii) when a platoon of vehicles follow a leader with a stop-and-go speed profile, the formation of congestion and subsequent dispersion are simulated. The results show that the model can reproduce the stochastic 2D trajectories of the vehicle and the marginal distribution of lateral movement. The proposed model can be used in both a simulation platform and a behavioral analysis of a human driver in traffic flow.

Exploring Behavioral Patterns of Lane Change Maneuvers for Human-Like Autonomous Driving

Due to the growing interest in automated driving, a deep understanding on the characteristics of human driving behavior is critical for human-like autonomous vehicles. Among various driving behaviors, lane change is the most important one for vehicle lateral driving safety. This study proposes an unsupervised method to extract and discover the behavioral patterns of lane change maneuvers for the purpose of exploring the composed behavioral patterns during lane change. This method involves two phases: Firstly, the lane change sequences will be segmented into blocks using time-series segmentation algorithms. Three segmentation algorithms were utilized in this study. In the second phase, the segments will be clustered to find the corresponding behavioral pattern of each segment. Two extended latent Dirichlet allocation (LDA) models were adopted to cluster the segments. The combination of different segmentation and clustering algorithms were evaluated and compared by employing entropy and perplexity as the evaluation criteria. Collected lane change data from naturalistic driving were applied to examine its effectiveness. The results show that this method could effectively mine descriptive behavioral patterns from lane change data. This study provides a promising data mining solution to facilitating deep and comprehensive understanding on driver lane change behaviors, which will promote the development of human-like autonomous vehicles.

An Implementation of Traffic Signs and Road Objects Detection Using Faster R-CNN

Traffic signs and road objects detection is significant issue for driver safety. It has become popular with the development of autonomous vehicles and driver-assistant systems. This study presents a real-time system that detects traffic signs and various objects in the driving environment with a camera. Faster R-CNN architecture was used as a detection method in this study. This architecture is a well-known two-stage approach for object detection. Dataset was created by collecting various images for training and testing of the model. The dataset consists of 1880 images containing traffic signs and objects collected from Turkey with the GTSRB dataset. These images were combined and divided into the training set and testing set with the ratio of 80/20. The model's training was carried out in the computer environment for 8.5 hours and approximately 10000 iterations. Experimental results show the real-time performance of Faster R-CNN for robustly traffic signs and objects detection.

A study on psychological determinants of users' autonomous vehicles adoption from anthropomorphism and UTAUT perspectives.

As the autonomous vehicles technology gradually enters the public eye, understanding consumers' psychological motivations for accepting autonomous vehicles is critical for the development of autonomous vehicles and society. Previously, researchers have explored the determinants of fully autonomous vehicles but the relevant research is far from enough. Moreover, the relationship between anthropomorphism and users' behavior has been ignored to a large extent. Therefore, this study aim to fill the gap by using anthropomorphism and the unified theory of acceptance and use of technology (UTAUT) to explore how system attributes (i.e., perceived anthropomorphism, perceived intelligence) and UTAUT attributes influence consumers' acceptance behavior. The data were collected via questionnaire survey conducted in Beijing, China, which can be a promising early adopter of AVs. Structural equation modeling was used to analyze the data. The results reveal that perceived anthropomorphism and perceived intelligence have a direct positive influence on the adoption of AVs; performance expectancy, effort expectancy, and facilitating conditions have an indirect positive influence on intention to adopt AVs. Also, this research contributes to the literature by enriching studies on psychological determinants of autonomous vehicles' adoption by taking an initial step to highlight anthropomorphism perceptions. This can provide managerial implications for policy-makers and businesses on how to effectively allocate resources to enhance autonomous vehicle adoption.

Self- Driving With Deep Reinforcement ( Survey )

We will focus concentrating on the scientific and technological challenges that the development of autonomous vehicles brings to a number of different manufacturers and research organizations (AVs). It is anticipated that automobiles operated by humans will continue to be a common sight on the roads for the foreseeable future; nevertheless, it is possible that autonomous vehicles may coexist in the same traffic environments. In order to keep traffic moving smoothly throughout the various sorts of mixed-use zones, autonomous vehicles (AVs) need to have driving rules and negotiation abilities that are analogous to those used by humans. In order to develop driving abilities that are comparable to those of humans, model-free deep reinforcement learning is used to simulate the actions of a skilled human driver. In this simulation, the difficulty of avoiding static obstacles on a two-lane roadway is investigated .

A Working Paper on the Regulatory Environment for Autonomous Vehicles

Whether it be 5, 10 or 20 years from now, fully self-driving care, also known as autonomous vehicles, are coming in the not-too-distant future. Yet despite the fact that driverless robo-taxis are already operating in some jurisdictions, there are still no federal regulations to guide their deployment. Autonomous vehicles will have – in in some places are already having – dramatic impacts on consumer mobility habits, the transportation and technology industries, and the economy as a whole. Yet despite the major disruption they represent, there are no national regulations guiding autonomous vehicle development. Instead, it is a state-by-state patchwork of regulations that guide the industry. The lack of standardization a uniform national regulatory framework would delay the development of autonomous vehicles and make it impossible for developers to achieve the the economies of scale standardization would bring – economies of scale that are critical for this new technology to be economically viable.

Joint Control of Power, Beamwidth, and Spacing for Platoon-Based Vehicular Cyber-Physical Systems

In a platoon-based vehicular cyber-physical system (VCPS), the traffic capacity and the fuel efficiency can be significantly improved by maintaining a small inter-vehicle distance (IVD). Meanwhile, the vehicle-to-infrastructure (V2I) communications provided by the roadside unit (RSU) are important for promoting the development of autonomous vehicles and reshaping the driving experience. However, as maintaining a small IVD inside the platoon may result in severe inter-beam interference, there is a tradeoff between the traffic capacity and the communication efficiency. In this paper, we formulate an optimization problem to minimize the IVD (a.k.a spacing) under the V2I quality-of-service (QoS) requirements by jointly controlling the transmit power, the beamwidth, and the IVD in a platoon-based VCPS. The traffic and channel uncertainties significantly complicate the analytical characterization of the impact of IVD, transmit power, and beamwidth on the QoS requirements. To solve the intractable challenging problem and capture the inherent timescale difference between the communication and vehicular systems, we propose a novel data-driven two-timescale reinforcement learning framework to control the beamforming and the spacing of the platoon at two timescales. With elaborately designed state, action, and reward, we develop a soft actor-critic based two-timescale deep reinforcement learning (DRL) algorithm to learn an effective policy for the control-communication co-design problem. Simulation results validate the effectiveness of spacing control and show that the proposed two-timescale DRL algorithm can significantly improve the traffic capacity and reduce the air drag coefficient while satisfying the QoS requirements of V2I communications.

Study of Human Comfort in Autonomous Vehicles Using Wearable Sensors

The rapid development of autonomous vehicles (AVs) has depicted a promising future of a safer and more efficient transportation system. To better induce this revolution, massive efforts have been spent on the technical competence of AVs. However, the human comfort in AVs has been an under-discussed yet important topic to the user acceptance of AVs. For the detection of human comfort, existing studies focus more on the physical influential factors of comfort such as sitting posture, vibration, and noise. With the introduction of AVs, psychological factors also have gained greater influence on human comfort. Despite existing studies of exploring correlations between human comfort and some physiological signals in automated driving contexts, there is few study on how human comfort level in AVs can be detected with these physiological signals. In this paper, we developed effective human comfort study approaches in autonomous vehicles with wearable sensors. We also proposed a machine learning based approach with adaptive feature selection to detect human comfort levels based on the wearable sensing data. The experimental results illustrated the effectiveness of the proposed approaches in studying human comfort in AVs.

A V2P Collision Risk Warning Method based on LSTM in IOV

With the evolution of communication networks, the Internet of Vehicles (IOV) continues to accelerate the safe and rapid development of autonomous vehicles. Vehicle-to-Pedestrian (V2P) communication is a key technology in autonomous vehicles and a potential solution to realize collaborative intelligence between vehicles and pedestrians. However, the existing V2P communication early warning system does not consider the uncertainty of pedestrian trajectory, and the determination of the collision area is limited to a single point, resulting in an inaccurate system judgment and limited improvement of traffic efficiency. This paper designs a new autonomous-oriented V2P communication network architecture and completes a V2P communication early warning system based on Long Range (LoRa). A V2P anticollision model is established, and a new V2P collision risk early warning method is proposed. In this method, danger index is introduced into the early warning of collision between pedestrian and vehicle. The long short-term memory (LSTM) artificial neural network is used to predict the pedestrian’s trajectory, so as to deduce the pedestrian-vehicle collision risk area when the pedestrian trajectory is uncertain. Meanwhile, the confidence probability is used to judge whether the pedestrian and vehicle are warned. The simulation shows that the V2P collision risk warning method proposed in this paper has good performance, which can accurately warn the pedestrian and vehicle under different vehicle speeds and Global Positioning System (GPS) positioning errors. At the same time, it reflects the characteristics of intelligence brought by using LSTM methods. Using the V2P communication early warning system based on LoRa to verify the experimental results show that when the GPS positioning accuracy is submeter level, the prediction accuracy is greater than 98%. The results of the proposed method show good performance and high detection rate.

A Review of Current and Historical Research Contributions to the Development of Ground Autonomous Vehicles for Agriculture

In this study, a comprehensive overview of the available autonomous ground platforms developed by universities and research groups that were specifically designed to handle agricultural tasks was performed. As cost reduction and safety improvements are two of the most critical aspects for farmers, the development of autonomous vehicles can be of major interest, especially for those applications that are lacking in terms of mechanization improvements. This review aimed to provide a literature evaluation of present and historical research contributions toward designing and prototyping agricultural ground unmanned vehicles. The review was motivated by the intent to disseminate to the scientific community the main features of the autonomous tractor named BOPS-1960, which was conceived in the 1960s at the Alma Mater Studiorum University of Bologna (UNIBO). Jointly, the main characteristics of the modern DEDALO unmanned ground vehicle (UGV) for orchard and vineyard operations that was designed recently were evaluated. The basic principles, technology and sensors used in the two UNIBO prototypes are described in detail, together with an analysis of UGVs for agriculture conceived in recent years by research centers all around the world.

Understanding Interest in Personal Ownership and Use of Autonomous Vehicles for Running Errands: An Exploration Using a Joint Model Incorporating Attitudinal Constructs

Transportation has been experiencing disruptive forces in recent years. One key disruption is the development of autonomous vehicles (AVs) that will be capable of navigating roadways on their own without the need for human presence in the vehicle. In a utopian scenario, AVs may enter the transportation landscape and foster a more sustainable and livable ecosystem with shared autonomous electric vehicles (SAEV) serving mobility needs and eliminating the need for private ownership. In a more dystopian scenario, AVs would be personally owned by households—enabling people to live farther away from destinations, inducing additional travel, and roaming roadways with zero occupants. Concerned with the potential deleterious effects of having personal AVs running errands autonomously, this paper aims to shed light on the level of interest in sending AVs to run errands and how that variable affects the intent to own an AV. Using data from a survey conducted in 2019 in four automobile-oriented metropolitan regions in the United States, the relationship is explored through a joint model system estimated using the generalized heterogeneous data model (GHDM) methodology. Results show that even after accounting for socio-economic and demographic variables as well as latent attitudinal constructs, the level of interest in having AVs run errands has a positive and significant effect on AV ownership intent. The findings point to the need for policies that would steer the entry and use of AVs in the marketplace in ways that avoid a dystopian future.

Autonomous Vehicle: Challenges and Implementation

Autonomous vehicles are the upcoming smart vehicles expected to be efficient, crash avoiding and driver less ideal urban vehicle. To attain this goal many organizations started research in this field to solve the real time problems faced during the development of autonomous vehicles. The first challenge in this phase is to customize the existing technology to develop autonomous vehicle with latest technologies. The future expectation is to develop fully automated vehicle without any flaws. This can be achieved by integrating various sensors, navigation software and various control algorithms, etc. However, the challenges and uncertainties are still vast because the application of autonomous driving environment involves not only the complex automotive technology but also the human-behavior, traffic management strategies, ethics, liability, policies etc., So as the result carmaker’s does not expect to commercially launch an autonomous vehicle in less time. From the technical perception the detection of the obstacles at high speed and a long distance is one of the highest difficulties. This survey paper shows an overview of the challenges faced by the autonomous vehicle and the technological implementations applied. Based on the research works of various researchers this literature survey is carried out and it can be used for future references.

Development of a solar-powered small autonomous surface vehicle for environmental measurements

Nowadays, the use of autonomous surface vehicles has increased in popularity. Autonomous surface vehicles can be adapted to perform water quality measurements, water contamination and hazards detection, garbage collection, aquatic animal population control, shore control, and other applications. This paper shows the novel construction of a small solar-powered autonomous surface vehicle. The vehicle was manufactured of composite materials of natural origin. It was equipped with an electric engine powered by ultralight photovoltaic panel, and a set of sensors and actuators used for autonomous operation and measurements of such factors, as pH, turbidity, purity, and oxygen content in water. The investigations carried out have been divided into three main parts. First, the geometry of the proposed autonomous surface vehicle was modelled using computational fluid dynamics and optimal shape of hull and deck was determined. Then, a dedicated construction of the photovoltaic panel was developed and tested under real conditions. The maximum power generated was 41.7 W when solar radiation was 850 W/m2. The matched power related to the area of 1 m2 was higher by 9.7% – 34.3% compared to other tested market-available panels. Finally, simulations in the TRNSYS software were carried out to determine the operational parameters of the developed autonomous surface vehicle under different operating scenarios. Simulations showed that the operation scenario of the autonomous surface vehicle significantly affects its primary energy consumption due to the energy supplied from the grid to recharge the system after each day of operation. The change in location can increase the primary energy savings, since among the investigated locations the primary energy saving ratio varied between about 30 and 50%. The analysis pointed out that the developed construction of the solar-powered small autonomous surface vehicle is a really promising solution and can be considered as an addition or replacement to the actual available water quality monitoring systems.

Multi-Access Edge Offloading Based on Physical Layer Security in C-V2X System

Due to the gap between calculation-intensive and latency-sensitive features of self-driving tasks and the vehicle's limited computation and storage resources, the development of autonomous vehicles is situated in a bottleneck. Fortunately, multi-access edge computing (MEC) architecture offers a promising measure to bridge this gap. It brings resources proximity to mobile devices at the network edge to support task offloading. Nonetheless, it introduces the problem of insecure wireless communication. Beyond that, the autonomous vehicle also has strict requirements on latency and energy consumption of task offloading technique owing to its safety-related and electricity-driven characteristics. Many scholars have studied the related problems widely. Nevertheless, few papers combine these three significant factors, namely information security, task latency and energy consumption. Thus, we put forward a mobility-ware task offloading scheme in cellular vehicle-to-everything system, which carefully considers these three requirements together. In this scheme, a time approximation algorithm is proposed to tackle the channel condition variations caused by the vehicle movement. Small-cell base station proactively sends artificial noise to debase wiretapper decoding capability and further prevent data exchanged between vehicles and MEC from being eavesdropped at the physical layer. Moreover, the optimal offloading proportion is found to minimize the weighted sum of latency and energy consumption. Experimental performances validate the effectiveness and feasibility of the proposed task offloading scheme.

The influence of seatback reclination on body kinematics during low-speed frontal impact

Safety tests are generally performed in accordance with strict procedures that involve the use of a carefully positioned dummy. Safety tests are rarely conducted with the occupant being in a non-standard position. In real life driving, occupants, especially passengers, choose a more comfortable reclined position, and in such a case the body dynamics are different from those investigated in the standard safety test. Furthermore, with the rapid development of autonomous vehicles, which provide the possibility of the position of occupants being unrestricted, the investigation of “out of position” body kinetics is becoming more important. The presented study aimed to evaluate body dynamics with various seatback reclinations. Body dynamics were verified by simulating frontal impact on a sled system with the use of a standard 50 percentile Hybrid III dummy. Points of interest located on the dummies head, neck, pelvis, and legs were traced, which allowed its trajectory to be evaluated. Additionally, the maximal extrusion and the time of motion were evaluated. It was found that the maximal extrusion in the longitude direction is the same for semi and fully reclined seats. Furthermore, a reclined seat causes head rotation, which can result in neck injuries.

Research On the Current Situation of Autonomous Vehicle Industry in China

Purpose – The purpose of this paper is exploring the challenges it faces in the development status of domestic autonomous vehicle, looks forward to its future, and hopes to promote the research and development in this field and provide reference for promoting the development of autonomous vehicle.
 Design/Methodology/Approach – Through the theoretical research on the development status of autonomous driving industry and national policies.
 Findings – This paper studies precision marketing and smart recommendation, creative advertising improves revenue structure, optimizing user experience, resolving copyright disputes.
 Research Implications – Affected by the COVID-19, the automobile industry as a whole is in a depressed state, and the future market competition will become increasingly fierce. Since the outbreak of the new crown disease, unmanned distribution vehicles have played an active role in epidemic prevention in many cities. The automobile industry is bound to complete the transformation from traditional manufacturing to intelligent manufacturing.

Inspection and identification of transmission line insulator breakdown based on deep learning using aerial images

The traditional methods of overhead power transmission line inspections are mostly unsuited as the height of transmission towers is too high and wide. Detection and inspection of insulators in aerial images with cluttered backgrounds is a challenging task for autonomous inspections. This manuscript mainly focuses on the development of autonomous Unmanned Aerial Vehicles (UAV/Quadcopter) that can hover over the transmission towers and capture images and videos by following pre-determined waypoints. To accomplish this, authors propose a new autonomous vision-based inspection that uses a Quadcopter as primary source of data, aerial images as the main source of information, and Deep Learning (DL) as the backbone analysis for inspection and focused on (i) insufficient training data, (ii) detection of insulators and their defects. A medium sized dataset of insulators for training and detection is created to overcome data insufficiency. The experimental results shows that the proposed deep learning architecture successfully identifies the anomalies of insulator such as, cracks, missing top caps and broken disk etc. The detection accuracy of the proposed deep learning algorithm can reach up to 93.5% with a detection speed of 58.2 frames/sec. The proposed DL algorithm has a promising potential towards smart inspection of insulators in power grids.

Bias-Learning-Based Model Predictive Controller Design for Reliable Path Tracking of Autonomous Vehicles Under Model and Environmental Uncertainty

Abstract Path tracking error control is an essential functionality in the development of autonomous vehicles to follow a planned trajectory. Significant path tracking errors could lead to a collision or even out of the control of the vehicle. Model-based control strategies have been developed to minimize the vehicle’s path tracking errors. However, the vehicle model may not truly represent the actual vehicle dynamics. Furthermore, the parameters employed in the vehicle dynamic model may not represent the actual operating conditions of the vehicle under environmental uncertainty. This paper proposes a real-time bias-learning method coupling with the model predictive control (MPC) to improve the fidelity of a baseline vehicle model with the aid of a few experiments (or virtual experiments) so that the path tracking error can be reduced in real-time operation. Gaussian process (GP) regression and recurrent neural network (RNN) are employed for bias-learning and their effectiveness are compared under different scenarios. GP regression learns non-linearity of the model bias through its nonlinear kernel function, whereas the RNN model formulates the bias as a linear combination of hidden nodes which capture the non-linearity of the model bias with a recurrent form. Results reveal that RNN is more effective for real-time learning of the nonlinear model bias than the classical GP regression and the proposed bias-learning model is able to improve the fidelity of a baseline vehicle dynamic model. Consequently, path tracking performance can be greatly improved under environmental uncertainty using the bias-learning-based MPC.

Driver Drowsiness Detection System

Abstract: Recently, in addition to autonomous vehicle technology research and development, machine learning methods have been used to predict a driver's condition and emotions in order to provide information that will improve road safety. A driver's condition can be estimated not only by basic characteristics such as gender, age, and driving experience, but also by a driver's facial expressions, bio-signals, and driving behaviours. Recent developments in video processing using machine learning have enabled images obtained from cameras to be analysed with high accuracy. Therefore, based on the relationship between facial features and a driver’s drowsy state, variables that reflect facial features have been established. In this paper, we proposed a method for extracting detailed features of the eyes, the mouth, and positions of the head using OpenCV and Dlib library in order to estimate a driver’s level of drowsiness. Keywords: Drowsiness, OpenCV, Dlib, facial features, video processing

Autonomous cars, development of modern transport technology

The article presents selected issues from the conducted research and analyzes in the field of the development of autonomous vehicles, carried out under the project: Polish road to Automation of Road Transport. The development of autonomous cars is a global challenge for the automotive industry. The creation of a fully autonomous vehicle capable of transporting passengers and goods without the participation of a driver has become a priority goal of automotive concerns. The huge costs incurred by automotive concerns for the development of autonomous vehicles prove the progressing automotive revolution, which will have a significant impact on the shape of future transport systems, transport based on autonomous transport. Very often in the literature there are overly optimistic forecasts for the development of autonomous vehicles, as well as overestimating the benefits they are to bring, without pointing to the costs and problems of their implementation and functioning in the real transport system. The publication presents forecasts for the development of autonomous vehicles as well as the related risks and advantages of their implementation. The factors that have a significant impact on the development of these vehicles are also outlined.