Self-driving Cars Research Articles

Route planning and obstacle avoidance using Q-Learning in MATLAB

Abstract. Recently, autonomous navigation has become important for many applications like self-driving cars, delivery drones, and robots. These domains are typically characterized by issues in planning an efficient route and object avoidance. Therefore, the paper describes work that utilizes Q-Learning as a means of reinforcement learning for autonomous navigation in terrain based on grid representations. Q-Learning allows an agent to acquire optimal signals in terms of strategy without the need for awareness of the surrounding environment. MATLAB was employed for the construction of the Q-Learning algorithm, as it was necessary to execute, evaluate, and simulate the algorithm in a controlled setting. In this study, the agent was tested in other environments of varying sizes (1010, 1515, and 2020) populated with randomly generated obstacles. Benchmark values such as average path length, number of collisions and convergence speed were observed as a measure of the agents performance. The analysis of the values obtained confirms that the agent is able to capture the shortest path from the starting point to the goal while avoiding the obstacles. Q-Learning is found to be flexible and effective in solving the presented navigation problems in this research. The future work would include the enhancement of the efficiency of the algorithm along with the applying of the system in dynamic and more complex scenarios, in order to solve practical problems.

Advanced obstacle recognition and trajectory prediction for autonomous vehicles using 3D CNN and GRU

Abstract. Recent years have seen a rapid advancement in new energy vehicle technology, which has increased consumer demand for autonomous driving systems. However, current self-driving cars still face many technical barriers. Obstacle detection and obstacle trajectory prediction in dynamic environments is one of the main difficulties of the technique. To improve self-driving cars' capacity for route planning and decision-making in dynamic contexts, this thesis proposes an obstacle identification and trajectory prediction model that combines a gated recurrent unit (GRU) network with a three-dimensional convolutional neural network (3D CNN). The integrated framework utilises the feature recognition capability of 3D CNN and the prediction capability of GRU to accurately predict dynamic obstacles, thereby improving the performance of the model in various types of environments (e.g., visual SLAM). Experimental results based on publicly available datasets (e.g., KITTI and Argoverse) show that the proposed method has significant advantages in dynamic obstacle identification and prediction as well as overall model operation efficiency. The findings show that the integrated framework is useful and successful in advancing the creation of self-driving automobiles.

Review of PID Control in Industrial Areas for Tobacco Production, Self-driving Car Technology and Robotic Arm Robots

Abstract. PID control is widely employed across diverse industries such as tobacco production, autonomous car technology, and mechanical arm robotics. This shows that PID control has a substantial developmental potential, so I choose the PID control as the topic. Since few people summarize its different applications in different industries, this study has the motivation to comprehensively summarize the applications of PID. This research provides an in-depth overview and analysis of PID control's varied applications in tobacco production, autonomous car technology, and mechanical arm robotics. It focuses on addressing practical challenges and exploring optimization strategies encountered in these fields. This article is summarized using the methods of the review. PID plays a crucial role in enhancing stability, precision, and response speed in these applications, thereby improving overall control system dynamic response performance. This study consolidated existing literature to offer insights into PID's effectiveness across these industrial sectors, shedding light on its impact and potential for further advancements in control technology.

Research on image recognition and processing application technology of unmanned vehicle based on deep learning

Abstract. Image recognition technology is critically important in various fields, including the rapidly advancing sector of autonomous vehicles. As one of the core components enabling driverless cars to perceive their environment and make informed decisions, image recognition has seen significant advancements due to deep learning. This paper focuses on the application of deep learning in image recognition for self-driving cars and explores its implications for the future of autonomous driving technology. To begin with, this paper examines the empirical evaluation of deep learning models in highway driving scenarios. By employing Convolutional Neural Networks (CNN), these models achieve high detection rates and superior accuracy in recognizing vehicles and lanes. The robustness of these models is tested under varying weather conditions and times, demonstrating their effectiveness compared to classical computer vision techniques. Next, the paper discusses radar-camera fusion technology, highlighting different fusion strategies such as data-level, feature-level, object-level, and hybrid-level. The findings suggest that while feature-level fusion excels in detecting small objects in complex scenes, hybrid-level fusion is optimal for diverse driving situations. This section provides valuable insights into the integration of multimodal data for improved object recognition and semantic segmentation. After discussing fusion technologies, the paper finally reviews the security challenges posed by adversarial attacks on deep learning-based unmanned systems.

A Comprehensive Review of Artificial Intelligence and Machine Learning in Control Theory

Abstract: Traditional control methods, such as proportional-integral-derivative (PID) controllers and linear-quadratic regulators (LQRs), have proven effective for linear and well-modeled systems. However, these methods often perform poorly in nonlinear, complex and dynamic environments. The paper aims to investigate the modern control systems by integrating artificial intelligence (AI) techniques, such as machine learning (ML), reinforcement learning (RL), deep learning, and fuzzy logic, to enhance their adaptive, robust, and predictive capabilities. And it reviews the literature and analyzes AI integration in control systems. The proposed strategies include supervised learning for trajectory optimization and fault detection, reinforcement learning for optimal control in dynamic environments, neural networks for complex nonlinear function approximation, and fuzzy logic for handling uncertainty and imprecise inputs. AI techniques significantly enhance the ability to tackle nonlinear problems and dynamic changes, demonstrating superior performance in applications like self-driving cars adapting to various road conditions and optimal energy distribution in smart grids. Despite the challenges of computational complexity, scalability, and the safety and reliability in the implementation of interpretable AI models, this paper suggests that hybrid approaches combining traditional control and AI techniques, along with the evolution of interpretable AI and convergence with quantum control, hold great promise for advancing AI-driven control systems.

A Root Cause Analysis of a Self-Driving Car Dragging a Pedestrian

A Root Cause Analysis of a Self-Driving Car Dragging a Pedestrian

자율주행 차량 사고 원인 분석을 통한 안전성 제고 방안

Although the convenience of drivers is maximizing due to the full-fledged self-driving service, complex problems are still emerging in various areas, including legal, ethical, and social aspects. As a result, social awareness and acceptability of self-driving cars have not been sufficiently formed, and efforts to improve stability are essential at this point. Therefore, in this study, only records of accidents while driving autonomously were selected among the accident data of autonomous vehicles provided by California DMV and the causes of accidents were analyzed for about 400 cases. As a result, main causes were verified as low recognition rate of parked vehicles, inability to grasp changes in vehicle size due to attachment of convenience goods, insufficient defense driving system, low recognition speed in special road conditions, etc. To mitigate these problems, it is expected to increase vehicle recognition rate, reconstruct safety area range, devise additional vehicle functionalities for defensive driving, and emphasize the importance of communication between self-driving cars and smart roads. These measures aim to enhance the safety of autonomous vehicles and contribute to increasing positive social perception and trust in autonomous driving.

Introduction to latent variable energy-based models: a path toward autonomous machine intelligence

Abstract Current automated systems have crucial limitations that need to be addressed before artificial intelligence can reach human-like levels and bring new technological revolutions. Among others, our societies still lack level-5 self-driving cars, domestic robots, and virtual assistants that learn reliable world models, reason, and plan complex action sequences. In these notes, we summarize the main ideas behind the architecture of autonomous intelligence of the future proposed by Yann LeCun. In particular, we introduce energy-based and latent variable models and combine their advantages in the building block of LeCun’s proposal, that is, in the hierarchical joint-embedding predictive architecture.

The Evolution of Edge Computing in a Data-Driven World

Edge computing is revolutionizing data processing and storage by bringing computational and data storage facilities closer to the data sources. This technique tackles the challenges faced by conventional clouds such as high latency and network congestion by employing the concept of data processed at the edge of the network in real-time. This paper discusses the edge computing paradigm and its evolution by covering the relevant concepts, benefits and trends that are changing within various industries. In particular, we assess the factors that make edge computing beneficial in terms of security, performance and efficiency concerns in smart cities, self-driving cars and internet of things (IoT) industries. This paper also reviews and reflects on the challenges of edge computing in terms of security, interoperability and scalability issues and how it can be overcome by integrating edge computing with fog computing, 5G and artificial intelligence (AI).

Ground Constrained 3D Lidar SLAM Design and Implementation

Abstract. Synchronized positioning and mapping techniques based on multi-line LIDAR are mainly used for mobile platforms that require high-precision positioning, such as UAVs and self-driving cars. However, multi-line LiDAR is limited by vertical field of view and vertical angular resolution, which can affect the vertical accuracy of attitude estimation. In this paper, we aim to extract the ground information in the environment and fuse it into a factor map to constrain the vertical accuracy of the attitude. Different from the traditional geometric information-based ground plane segmentation, this paper utilizes point cloud distribution information to extract the ground cloud and uses high-confidence ground features to constrain the attitude drift by considering their distribution weights. Finally, the laser range factor, ground constraint factor and closed-loop factor are integrated under the factor graph optimization framework. The effectiveness of the method is validated on a self-constructed dataset.

How the technologies behind self-driving cars, social networks, ChatGPT, and DALL-E2 are changing structural biology.

The performance of deep Neural Networks (NNs) in the text (ChatGPT) and image (DALL-E2) domains has attracted worldwide attention. Convolutional NNs (CNNs), Large Language Models (LLMs), Denoising Diffusion Probabilistic Models (DDPMs)/Noise Conditional Score Networks (NCSNs), and Graph NNs (GNNs) have impacted computer vision, language editing and translation, automated conversation, image generation, and social network management. Proteins can be viewed as texts written with the alphabet of amino acids, as images, or as graphs of interacting residues. Each of these perspectives suggests the use of tools from a different area of deep learning for protein structural biology. Here, I review how CNNs, LLMs, DDPMs/NCSNs, and GNNs have led to major advances in protein structure prediction, inverse folding, protein design, and small molecule design. This review is primarily intended as a deep learning primer for practicing experimental structural biologists. However, extensive references to the deep learning literature should also make it relevant to readers who have a background in machine learning, physics or statistics, and an interest in protein structural biology.

How Does Talking with a Human-like Machine in a Self-Driving Car Affect your Experience? A Mixed-Method Approach

This study investigates the impact of human-like machines (HLMs) on the user experience (UX) of young adults during voice interactions between drivers and autonomous vehicles. A mixed-method approach was employed to evaluate three voice agents with varying levels of anthropomorphism: a machine voice without humanized speech strategies (Agent A), a human voice without humanized speech strategies (Agent B), and a human voice with humanized speech strategies (Agent C). A total of 30 participants were invited to interact with the agents in a simulated driving scenario. Quantitative measures were employed to assess intimacy, trust, intention to use, perceived safety, and perceived anthropomorphism based on a 7-point Likert scale, while qualitative interviews were conducted to gain deeper insights. The results demonstrate that increased anthropomorphism enhances perceived anthropomorphism (from 2.77 for Agent A to 5.01 for Agent C) and intimacy (from 2.47 for Agent A to 4.52 for Agent C) but does not significantly affect trust or perceived safety. The intention to use was higher for Agents A and C (4.56 and 4.43, respectively) in comparison to Agent B (3.88). This suggests that there is a complex relationship between voice characteristics and UX dimensions. The findings of this study highlight the importance of balancing emotional engagement and functional efficiency in the design of voice agents for autonomous vehicles.

Proposed method for against adversarial images based on ResNet architecture

Our world is becoming increasingly automated due to the application of deep learning/machine learning models to systems, but these systems are vulnerable to adversarial attacks, which create deceptive data to trick them. Without proper defenses, attackers can exploit deep learning systems in facial recognition, self-driving cars, and social media filters. Research on adversarial image generation and methods to against attacks is important. This paper proposes employing the ResNet architecture with adversarial training to against adversarial images. The model is tested on a Hybrid CIFAR-10 dataset, which is designed to improve robustness and accuracy by incorporating GAN-generated images. The proposed model achieves an accuracy of over 95%, which is better than three other state-of-the-art architectures VGG19_bn, ShuffleNetV2, and RepVGG_a2.

How AI Systems Can Be Blameworthy

AbstractAI systems, like self-driving cars, healthcare robots, or Autonomous Weapon Systems, already play an increasingly important role in our lives and will do so to an even greater extent in the near future. This raises a fundamental philosophical question: who is morally responsible when such systems cause unjustified harm? In the paper, we argue for the admittedly surprising claim that some of these systems can themselves be morally responsible for their conduct in an important and everyday sense of the term—the attributability sense. More specifically, relying on work by Nomy Arpaly and Timothy Schroeder (In Praise of Desire, OUP 2014), we propose that the behavior of these systems can manifest their ‘quality of will’ and thus be regarded as something they can be blameworthy for. We develop this position in detail, justify some of its crucial presuppositions, and defend it against potential objections.

Fuzzy Petri Nets for Traffic Node Reliability.

Self-driving cars are one of the main areas of research today, but it has to be acknowledged that the information from the sensors (the perceptron) is a huge amount of data, which is now unmanageable even when projected onto a single traffic junction. In the case of self-driving, the nodes have to be sequenced and organized according to the planned route. A self-driving car in Hungary would have to be able to interpret more than 70,000 traffic junctions to be able to drive all over the country. Besides the huge amount of data, another problem is the issue of validation and verification. For self-driving cars, this implies a level of complexity using traditional methods that calls into question the economics of the already existing system. Fuzzy Petri nets provide an alternative solution to both problems. They allow us to obtain a model that accurately describes the reliability of a node through its dynamics, which is essential in perception since the more reliable a node is, the smaller the deep learning mesh required. In this paper, we outline the analysis of a traffic node's safety using Petri nets and fuzzy analysis to gain information on the reliability of the node, which is essential for the modeling of self-driving cars, due to the deep learning model of perception. The reliability of the dynamics of the node is determined by using the modified fuzzy Petri net procedure. The need for a fuzzy extension of the Petri net was developed by knowledge of real traffic databases.

HUMAN RESOURCE MANAGEMENT AND ARTIFICIAL INTELLIGENCE: TRANSFORMATIVE EFFECTS AND FUTURE PROSPECTS

Artificial intelligence (AI) is technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy. Ancient civilizations contained tales of intelligent robots, such as Greek and Chinese cultures. AI as a modern phenomenon which took shape in the mid-20th century with the advent of digital computers. The first breakthrough, published in 1950, was Alan Turing's paper, ‘Computing Machinery and Intelligence. AI can make provide detailed solution to users and experts. They can act independently, replacing the need for human intelligence or intervention (a classic example being a self-driving car). Many sectors like finance, healthcare, to human resource management (HRM), rely on AI technologies. AI's integration in HRM has brought about great change in conventional HR practices used to measure efficiency, accuracy, strategic decision-making, etc. AI technologies help to provide solution to automation of repetitive tasks, analysis of sprawling volumes of data and insights facilitating decision-making. Through this research paper we try to understand the AI's effects on HR processes, in terms of recruitment, onboarding, performance appraisal, training and development, employee engagement, workforce planning; understanding the benefits and cons of implementing AI in HRM; ethical considerations and potential biases in AI algorithms relevant to HR practices; analysis of emergent AI trends with respect to HRM. Trends in AI in HR tend towards a most promising future, presenting several innovations. Accordingly, HR professionals will be expected to acquire competency with regards to new roles and responsibilities to avail AI in enhancing strategic planning, ethical oversight, and data-driven decision-making.

Enhanced Visual SLAM for Collision-Free Driving with Lightweight Autonomous Cars.

The paper presents a vision-based obstacle avoidance strategy for lightweight self-driving cars that can be run on a CPU-only device using a single RGB-D camera. The method consists of two steps: visual perception and path planning. The visual perception part uses ORBSLAM3 enhanced with optical flow to estimate the car's poses and extract rich texture information from the scene. In the path planning phase, the proposed method employs a method combining a control Lyapunov function and control barrier function in the form of a quadratic program (CLF-CBF-QP) together with an obstacle shape reconstruction process (SRP) to plan safe and stable trajectories. To validate the performance and robustness of the proposed method, simulation experiments were conducted with a car in various complex indoor environments using the Gazebo simulation environment. The proposed method can effectively avoid obstacles in the scenes. The proposed algorithm outperforms benchmark algorithms in achieving more stable and shorter trajectories across multiple simulated scenes.

Improved Multi-Sensor Fusion Dynamic Odometry Based on Neural Networks

High-precision simultaneous localization and mapping (SLAM) in dynamic real-world environments plays a crucial role in autonomous robot navigation, self-driving cars, and drone control. To address this dynamic localization issue, in this paper, a dynamic odometry method is proposed based on FAST-LIVO, a fast LiDAR (light detection and ranging)–inertial–visual odometry system, integrating neural networks with laser, camera, and inertial measurement unit modalities. The method first constructs visual–inertial and LiDAR–inertial odometry subsystems. Then, a lightweight neural network is used to remove dynamic elements from the visual part, and dynamic clustering is applied to the LiDAR part to eliminate dynamic environments, ensuring the reliability of the remaining environmental data. Validation of the datasets shows that the proposed multi-sensor fusion dynamic odometry can achieve high-precision pose estimation in complex dynamic environments with high continuity, reliability, and dynamic robustness.

Revolutionizing Roads: The Impact of Autonomous Vehicles on Future Travel

This project focuses on developing an autonomous self-driving car by integrating advanced algorithms with cutting-edge hardware. The vehicle is designed to perceive its environment, make informed decisions, and navigate complex traffic scenarios autonomously by leveraging machine learning and computer vision technologies. Essential methods include using deep learning algorithms for object detection and classification, sensor fusion techniques to combine data from cameras, LiDAR, and radar, and reinforcement learning for decision-making in dynamic environments. The project emphasizes the critical role of AI in enabling the car to interpret and respond to dynamic road conditions while ensuring reliable performance through rigorous testing and validation in simulated and real-world scenarios. Initial results from testing indicate that the car can successfully navigate complex traffic situations, such as merging onto highways and responding to pedestrians, with a high degree of accuracy. By pushing the boundaries of autonomous vehicle technology, this research aspires to contribute to the ongoing evolution of self-driving cars and their potential to revolutionize mobility. With a focus on safety, efficiency, and innovation, this endeavor represents a significant step forward in developing autonomous vehicles, offering new possibilities for the future of transportation. The project promises to improve travel convenience and significantly reduce traffic accidents, ultimately contributing to a safer and more efficient road system.

Object and Pedestrian Detection on Road in Foggy Weather Conditions by Hyperparameterized YOLOv8 Model

Connected cooperative and automated (CAM) vehicles and self-driving cars need to achieve robust and accurate environment understanding. With this aim, they are usually equipped with sensors and adopt multiple sensing strategies, also fused among them to exploit their complementary properties. In recent years, artificial intelligence such as machine learning- and deep learning-based approaches have been applied for object and pedestrian detection and prediction reliability quantification. This paper proposes a procedure based on the YOLOv8 (You Only Look Once) method to discover objects on the roads such as cars, traffic lights, pedestrians and street signs in foggy weather conditions. In particular, YOLOv8 is a recent release of YOLO, a popular neural network model used for object detection and image classification. The obtained model is applied to a dataset including about 4000 foggy road images and the object detection accuracy is improved by changing hyperparameters such as epochs, batch size and augmentation methods. To achieve good accuracy and few errors in detecting objects in the images, the hyperparameters are optimized by four different methods, and different metrics are considered, namely accuracy factor, precision, recall, precision–recall and loss.