Stability Of Autonomous Vehicle Research Articles

Improving the Performance of Autonomous Driving through Deep Reinforcement Learning

Reinforcement learning (RL) is revolutionizing the artificial intelligence (AI) domain and significantly aiding in building autonomous systems with a higher level comprehension of the world as we observe it. Deep learning (DL) facilitates RL to scale and resolve previously intractable problems, for instance, allowing supervision principles designed for robots to be acquired directly from visual data, developing video game proficiency from pixel-level information, etc. Recent research shows that RL algorithms help represent problems dealing with high-dimensional, unprocessed data input and can have successful applications in computer vision, pattern identification, natural language analysis, and speech parsing. This research paper focuses on training a simulation model of a car to navigate autonomously on a racetrack using RL. The study explores several fundamental algorithms in Deep RL, namely Proximal Policy Optimization (PPO), Deep Q-network (DQN), and Deep Deterministic Policy Gradient (DDPG). The paper documents a comparative analysis of these three prominent algorithms—based on their speed, accuracy, and overall performance. After a thorough evaluation, the research indicates that the DQN surpassed the other existing algorithms. This study further examined the performance of the DQN with and without ε-decay and observed that the DQN with ε-decay is better suited for our objective and is significantly more stable than its non ε-decay counterpart. The findings from this research could assist in improving the performance and stability of autonomous vehicles using the DQN with ε -decay. It concludes by discussing the fine-tuning of the model for future real-world applications and the potential research areas within the field of autonomous driving.

A model of lane-changing intention induced by deceleration frequency in an automatic driving environment

Achieving high-level, anthropomorphic lane-changing control for autonomous vehicles is an important method for promoting the safe and efficient operation of autonomous vehicles. According to existing lane-changing models, the important phenomenon that frequent acceleration and deceleration cause a poor riding experience and thus induce the intention of lane-changing needs to be given more attention. In this study, a discretionary lane-changing control method for autonomous vehicles is designed based on anti-interference ability (AIA, where AIA is the number of decelerations per unit time of an autonomous vehicle) to extract the traffic flow variation characteristics of autonomous vehicles with AIA and to determine methods to improve the driving efficiency and stability of autonomous vehicles. AIA is introduced to the lane-changing decision-making process of autonomous vehicles to simulate the phenomenon of lane-changing intention caused by poor riding experience. A control model that can evaluate the lane-changing conditions is then proposed. NetLogo software was selected to construct a one-way, two-lane scenario for autonomous vehicle operation, and the impacts of the proposed decision-making method in each scenario were tested. The experimental results show that autonomous vehicles are more likely to produce frequent deceleration behaviors under large-traffic and small-space operating conditions. A traffic flow composed of autonomous vehicles that use AIA to induce lane-changing intentions has a more obvious phase transition process. An autonomous vehicle whose AIA induces lane-changing intention can improve the speed and stability for a certain traffic volume (a saturation of less than approximately 0.45). Although lane-changing behavior breaks the stable state of two-lane traffic flow, it will also successfully suppress the aggravation of traffic congestion. Lane-changing has an impact on the local oscillation of traffic flow, and a stable traffic flow operation state and reduction in the AIA clearing frequency are mutually reinforcing.

Path-Tracking Considering Yaw Stability With Passivity-Based Control for Autonomous Vehicles

In this paper, a new passivity-based control approach is designed to improve the robustness and stability of autonomous vehicles in performing path-tracking tasks. A port-Hamiltonian model is a special geometric structure and provides a systematic and insightful framework to describe many physical systems from an energy perspective. The passivity-based control establishes a new and simple structure to achieve the path-tracking task by the port-Hamiltonian structure. Firstly, the path-tracking error system is transformed into a port-Hamiltonian system with the perturbation. The energy-shaping method is utilized to ensure the asymptotic stability, and the stability proof of the closed-loop, path-tracking error system is given. The proposed real-time, passivity-based controller is directly dependent on the passive outputs, which is robust to parameter uncertainties caused by load changes. Moreover, the control performance of steering control will be degraded when the steering angle is saturated. The yaw-moment control is introduced to enhance the driving safety. In the wheel torque distribution, an optimization-based control allocation is designed to minimize the sum of tire loads and to make them more evenly distributed to each tire. Finally, simulations are implemented in MATLAB/Simulink and CarSim co-simulation to demonstrate the effectiveness of the designed strategy under different driving conditions.

Evaluation on Braking Stability of Autonomous Vehicles Running along Curved Sections Based on Asphalt Pavement Adhesion Properties

As the main objective influencing factor on the brake safety of autonomous vehicles, pavement texture information is directly related to road surface antiskid performance. However, in the brake system of autonomous vehicles, the influence of road surface adhesion characteristics on braking stability is seldomly considered. To study the braking stability of autonomous vehicles on curved sections under different road conditions, the advanced close-range photogrammetry system was utilized to extract the road surface texture information. Thereafter, the power spectral density (PSD) of the road surface was calculated by MATLAB to obtain the pavement adhesion coefficient curves based on the Persson friction theory model under different road conditions. Considering the pavement adhesion characteristics, the braking model of autonomous vehicles was built in Simulink, and then, the braking performance on curved sections was analyzed with CarSim/Simulink cosimulation. The results indicate that, according to the adhesion coefficient of different asphalt pavement types under different road conditions, the ranking order is open-grade friction course (OGFC) > stone matrix asphalt (SMA) > dense-graded asphalt concrete (AC). In addition, both the maximum lateral offset distances and the maximum lateral forces of the tires decrease as the curve radius gradually increases under different road conditions. It can also be found that there is a relatively uniform vertical forces distribution of the tire when the curve radius is no less than 100 m, and the limit speed of the vehicle varies parabolically with increasing in curve radius. Compared with dry road, the reduction of vehicle braking deceleration is more significant and the yaw rate is greater on wet road. Last but not least, the braking comfort with a radius of 200 m is the best according to the comfort index (CI) in International Standard ISO, in which the comfort level can be sorted into six levels.

Coordination control of active front steering and direct yaw moment control based on stability judgment for AVs stability enhancement

A coordination control strategy based on stability judgment is presented for autonomous vehicles (AVs) aiming to enhance the handling and stability performance. Firstly, the stability judgment scheme is used to evaluate the real-time stability level of vehicles based on the Self-Organizing Feature Map (SOFM) neural network and K-Means algorithm. Secondly, a coordination controller of active front steering (AFS) and direct yaw moment control (DYC) is designed to track the desired vehicle motion. To enhance the handling and stability of AVs, the weights of AFS and DYC controllers are adaptively adjusted according to the vehicle stability level. Finally, the effectiveness of the proposed method is verified in co-simulation environment of CarSim and Simulink, and a rapid control prototyping test is implemented to evaluate the feasibility and robustness. The results indicate that the stability judgment scheme and coordination control strategy for AVs can not only satisfy the requirements of path tracking accuracy but also enhance the handling and stability performance.

Design and development of sensor based Automatic steering control system for automobiles

Drowsy driving is the major problem seen throughout the world which usually happens when the driver has not slept enough, and this state of mind may lead to accidents. Above being the case, the project helps driver to keep him alert by giving an alarm and to minimize accidents due to driver error, distractions and drowsiness. The research derives, implements, tunes and compares selected path tracking methods for controlling a car-like robot along a predetermined path. The model mainly aims on using active steering control to increase safety and to reduce both accidents and drivers work load. The objective of the project is to design and develop a steering control system for the path tracking of autonomous vehicle & to understand the stability of the vehicles on curved road. The steering controller controls the steering actuator to follow the desired steering angle. A servo motor is installed to control the steering handle, and it can transmit the steering force using a belt and pulley. The developed mechanism is a steering controller that is applied to a micro controller. However, because of a dead band, the path tracking performance and stability of autonomous vehicles are reduced. To overcome this, steering system and programme is developed. As a result of the steering system and programme, the performance and stability are improved. GPS shows the geographical coordinates which specifies any given location on the earth surface as latitude and longitude. Ultrasonic sensor used for detecting the obstacles. A magnetometer compass shows the quadrants. Arduino UNO board is used like a mother board in the project. Finally, the project augments the literature with a comprehensive collection of important path tracking ideas, a guide to their implementations and, most importantly, an independent and realistic comparison of the performance of these various approaches.

A deep learning platooning-based video information-sharing Internet of Things framework for autonomous driving systems

To enhance the safety and stability of autonomous vehicles, we present a deep learning platooning-based video information-sharing Internet of Things framework in this study. The proposed Internet of Things framework incorporates concepts and mechanisms from several domains of computer science, such as computer vision, artificial intelligence, sensor technology, and communication technology. The information captured by camera, such as road edges, traffic lights, and zebra lines, is highlighted using computer vision. The semantics of highlighted information is recognized by artificial intelligence. Sensors provide information on the direction and distance of obstacles, as well as their speed and moving direction. The communication technology is applied to share the information among the vehicles. Since vehicles have high probability to encounter accidents in congested locations, the proposed system enables vehicles to perform self-positioning with other vehicles in a certain range to reinforce their safety and stability. The empirical evaluation shows the viability and efficacy of the proposed system in such situations. Moreover, the collision time is decreased considerably compared with that when using traditional systems.

Analysis of Take-over Time and Stabilization of Autonomous Vehicle Using a Driving Simulator

Analysis of Take-over Time and Stabilization of Autonomous Vehicle Using a Driving Simulator

Nonlinear Coordinated Steering and Braking Control of Vision-Based Autonomous Vehicles in Emergency Obstacle Avoidance

This paper discusses dynamic control design for automated driving of vision-based autonomous vehicles, with a special focus on the coordinated steering and braking control in emergency obstacle avoidance. An autonomous vehicle is a complex multi-input and multi-output (MIMO) system, which possesses the features of parameter uncertainties and strong nonlinearities, and the coupled phenomena of longitudinal and lateral dynamics are evident in a combined cornering and braking maneuver. In this work, an effective coordinated control system for automated driving is proposed to deal with these coupled and nonlinear features and reject the disturbances. First, a vision algorithm is constructed to detect the reference path and provide the local location information between vehicles and reference path in real time. Then, a novel coordinated steering and braking control strategy is proposed based on the nonlinear backstepping control theory and the adaptive fuzzy sliding-mode control technique, and the asymptotic convergence of the proposed coordinated control system is proven by the Lyapunov theory. Finally, experimental tests manifest that the proposed control strategy possesses favorable tracking performance and enhances the riding comfort and stability of autonomous vehicles.

Development of Steering Control System for Autonomous Vehicle Using Geometry-Based Path Tracking Algorithm

In this paper, a steering control system for the path tracking of autonomous vehicles is described. The steering control system consists of a path tracker and primitive driver. The path tracker generates the desired steering angle by using the look-ahead distance, vehicle heading, and a lateral offset. A method for applying an autonomous vehicle to path tracking is an advanced pure pursuit method that can reduce cutting corners, which is a weakness of the pure pursuit method. The steering controller controls the steering actuator to follow the desired steering angle. A servo motor is installed to control the steering handle, and it can transmit the steering force using a belt and pulley. We designed a steering controller that is applied to a proportional integral differential controller. However, because of a dead band, the path tracking performance and stability of autonomous vehicles are reduced. To overcome the dead band, a dead band compensator was developed. As a result of the compensator, the path tracking performance and stability are improved.