Autonomous Vehicle Simulation Research Articles

Development, Simulation and Visualization of Autonomous Farm Vehicles

Abstract: A Research in Autonomous Agricultural Vehicle Technology is presented. With the coming of the Smart Farming Revolution the development of more complicated machines has begun. Agricultural machines are made with the capability of driving themselves, using GPS maps and electronic sensors, an approach that will lead to the development of futuristic Precise Autonomous Farming Systems. The basic issues to be addressed are autonomous path planning, obstacle detection and avoidance for which various sensors will be employed such as GPS, laser-based sensors, ultrasonic sensors, IMU sensors and Camera sensors whose data has to be constantly monitored using Dead Reckoning algorithm and constant Pose Estimation. In these sensors, Camera, Ultrasonic, Laser-Based Sensors are used as Detection sensors while the IMU and GPS sensors are used as Localization sensors. Global Avoidance Subsystem which examines the whole environment of an autonomous vehicle mission-level path planner that pre-plans paths around all known obstacles. While Local Avoidance Subsystem/Reactive planner consists of an obstacle filter and an obstacle avoidance algorithm and re-plans a short way for the vehicle to navigate. With the ROS framework and all sensors data, a vehicle can guide itself through an unknown environment. This will help the farmer to undercut the labour shortage and improve the precision in farming which in turn improves the output to meet the growing demands

Autonomous Driving System Architecture with Integrated ROS2 and Adaptive AUTOSAR

In the automotive industry, research is now underway to apply Adaptive Automotive Open System Architecture (AUTOSAR) to the development of next-generation mobility, such as autonomous driving and connected cars. However, research on autonomous driving is being predominantly conducted on the robotics platform ROS2 (Robot Operating System 2). This demonstrates a considerable distance between autonomous driving research and its application in actual vehicles. To bridge this gap, interoperability that leverages the strengths of the Adaptive AUTOSAR and ROS2 platforms and compensates for their weaknesses is required. Therefore, this study proposes an architecture for interoperability between the two platforms, named Autonomous Driving System with Integrated ROS2 and Adaptive AUTOSAR (ASIRA). The proposed architecture enables communication between each of the two platforms through the ROS2 SOME/IP Bridge and allows for the necessary data exchange. It validates them in autonomous driving scenarios and goes beyond vehicle development, testing, and prototyping to exploit the advantages of each platform. Additionally, the simulation of autonomous vehicles within the ASIRA architecture is demonstrated by interoperating the ROS2 representative open-source autonomous driving project, Autoware, with the Adaptive AUTOSAR simulator. This study contributes to the assimilation of ROS2 into the automotive industry and its application in real vehicles by linking ROS2 and Adaptive AUTOSAR.

Research on Behavioral Decision at an Unsignalized Roundabout for Automatic Driving Based on Proximal Policy Optimization Algorithm

Unsignalized roundabouts have a significant impact on traffic flow and vehicle safety. To address the challenge of autonomous vehicles passing through roundabouts with low penetration, improve their efficiency, and ensure safety and stability, we propose the proximal policy optimization (PPO) algorithm to enhance decision-making behavior. We develop an optimization-based behavioral choice model for autonomous vehicles that incorporates gap acceptance theory and deep reinforcement learning using the PPO algorithm. Additionally, we employ the CoordConv network to establish an aerial view for spatial perception information gathering. Furthermore, a dynamic multi-objective reward mechanism is introduced to maximize the PPO algorithm’s reward pool function while quantifying environmental factors. Through simulation experiments, we demonstrate that our optimized PPO algorithm significantly improves training efficiency by enhancing the reward value function by 2.85%, 7.17%, and 19.58% in scenarios with 20, 100, and 200 social vehicles, respectively, compared to the PPO+CCMR algorithm. The effectiveness of simulation training also increases by 11.1%, 13.8%, and 7.4%. Moreover, there is a reduction in crossing time by 2.37%, 2.62%, and 13.96%. Our optimized PPO algorithm enhances path selection during autonomous vehicle simulation training as they tend to drive in the inner ring over time; however, the influence of social vehicles on path selection diminishes as their quantity increases. The safety of autonomous vehicles remains largely unaffected by our optimized PPO algorithm.

How Do Humanlike Behaviors of Connected Autonomous Vehicles Affect Traffic Conditions in Mixed Traffic?

Different methodologies are being used to study the effects of autonomous vehicles (AVs) in mixed traffic to exhibit the interactions between autonomous and human-driven vehicles (HVs). Microscopic simulation tools are popular in such an assessment, as they offer the possibility to experiment in economical, robust, and optimistic ways. A lack of reliable real-world data (also known as natural data) to calibrate and evaluate the connected autonomous vehicle (CAV) simulation model is a major challenge. To deal with this situation, one interesting methodology could be to deal with the CAVs as conventional human-driven vehicles and predict their possible characteristics based on the simulation inputs. The conventional human-driven vehicles from the real world, in this methodology, come to act as a benchmark to offer the measure of effectiveness (MoE) for the calibration and validation. For the three most common driving behaviors, a sensitivity analysis of the behaviors of AVs and an effective assessment of CAVs in a mixed traffic environment were performed to explore the humanlike behaviors of the autonomous technology. The findings show that, up to a certain point, which is directly related to the quantity of interacting vehicles, the impact of CAVs is typically favorable. This study validates the approach and supports past studies by showing that CAVs perform better than AVs in terms of their traffic performance and safety aspects. On top of that, the sensitivity analysis shows that enhancements in the technology are required to obtain the maximum advantages.

Transparent integration of autonomous vehicles simulation tools with a data-centric middleware

Transparent integration of autonomous vehicles simulation tools with a data-centric middleware

Models, algorithms, and equilibrium conditions for the simulation of autonomous vehicles in exclusive and mixed traffic

This paper proposes models and algorithms for the estimation of traffic flows generated by the centralised management of autonomous vehicles under exclusive and mixed traffic conditions. It is assumed that the autonomous vehicle manager wants to obtain the system optimum that minimises the total travel time on the network, of autonomous vehicles in the case of exclusive traffic, and of all flows in the case of mixed traffic. In the case of mixed traffic, the drivers of the human-guided vehicles choose their routes from a user-optimal perspective; in this case, a fixed-point problem has to be solved, which represents a state of equilibrium between the flows of human-guided vehicles, the flows of autonomous vehicles, which follow paths aimed at the system optimum, and the link costs. The proposed models and algorithms are tested on a small network with a single origin-destination pair, and on the Sioux Falls network, with different levels of congestion and different percentages of autonomous vehicles. The results show the validity of the proposed methods and the impact of autonomous vehicles on network performance. The proposed static/equilibrium approach is necessary to evaluate transport planning, design or policy interventions that include the presence of autonomous vehicles in traffic flows.

The Design and Simulation of Autonomous Agricultural Vehicle for Orchards

In recent years, agricultural producers have faced challenges due to the uncertainty of labor force access, growing demand for safe, accessible, and high-quality agricultural products, high competition with international producers, and the need to reduce their carbon footprint. To continue their competitive and profitable production, producers must invest in technology and increase efficiency. Autonomous agricultural vehicles are crucial for autonomous processes in orchards, increasing productivity, collecting data for decision-making, reducing operating costs, and carbon footprint. This study focuses on the design and simulation of an autonomous vehicle for orchards. The autonomous vehicle can map the orchards using data from odometry and light detection and ranging (LIDAR) sensors by utilizing Simultaneous Localization and Mapping (SLAM) algorithm, accurately determine its position using the Adaptive Monte Carlo Localization method, and avoid obstacles using the dynamic window approach algorithm. The autonomous vehicle is an original design for netted orchards where GPS cannot work properly and is fully autonomous, requiring no external GPS data. It is expected to provide higher efficiency by reducing environmental pollution, operating expenses, and labor force in practice. The success of the mapping and localization application depends on the update frequency of the position and the number of particles used for localization. A path-planning application was developed to reach the desired position from the autonomous agricultural vehicle's current position on the map. The Dijkstra Algorithm was used for path planning, with the Dynamic Window Approach allowing the robot to escape obstacle. The simulation studies achieved the lowest position error when the vehicle's position was updated at intervals of 2 cm, and a minimum of 500 and a maximum of 2000 particles were used. While the vehicle was moving on a straight obstacle row in the simulation environment, an average localization error of 2.1 cm was obtained. This error is convenient as it enables seamless navigation between tree rows without any collision.

Quantifying automated vehicle benefits in reducing driving stress: a simulation experiment approach

Driving is a stressful activity because of the mental workload required to maneuver a vehicle in certain travel contexts, such as congested traffic, multi-modal networks requiring complex interaction with surrounding vehicles, and aggressive driving. Autonomous vehicles (AVs), on the other hand, can reduce the mental workload by performing most of the driving tasks and providing users with a comfortable ride. This study develops a pathway model to relate different health determinants, including travel reliability, safety, driving comfort, and value of time, to Autonomous vehicles driving and studies their impact on the value of driving stress. A case study example of Autonomous vehicles simulation is used to determine the impact of these health determinants. The value of driving stress in Autonomous vehicles is estimated as a function of the value of these individual health determinants. The results show that the perception of safe or unsafe driving in Autonomous vehicles is the most important factor in changing the perception of driving stress in Autonomous vehicles. Similarly, perceptions of comfortable driving in Autonomous vehicles and reduced workload with a higher value of time also reduce driving stress in Autonomous vehicles. These results allow Autonomous vehicles adoption models to explicitly consider driving stress reduction as a benefit and can improve understanding of Autonomous vehicles adoption, which may require quantitative analysis of underlying motivating benefits, including driving stress reduction.

Design and Implementation of Autonomous Underwater Vehicle Simulation System Based on MOOS and Unreal Engine

An effective autonomous underwater vehicle (AUV) simulation system can greatly improve development efficiency and reduce the cost and risk of actual equipment operation. In this paper, a comprehensive simulation system is developed using Mission Oriented Operating Suite (MOOS) and Unreal Engine 4 (UE4). The former provides an open-source development framework and application components, which are widely used in the field of underwater robots. The latter is a well-known game development engine that has realistic effects and various plugins. As far as we know, there are few simulation systems based on MOOS and UE4. Moreover, a modular architecture for the simulation system is designed and implemented in the paper. The developed system consists of four subsystems, including the simulation editing and controlling subsystem, shore-based operation subsystem, simulation engine subsystem, and three-dimensional (3D) visualization subsystem. These subsystems have modular characteristics and independent functions and communicate with each other through network sockets. Through the system, users can customize simulation scenarios, send missions and control simulation progress from the graphic user interface. And the system can be used in operation training, software testing, algorithm verification of AUV, etc.

Modeling Driver’s Real-Time Confidence in Autonomous Vehicles

Autonomous vehicle technology has developed at an unprecedented rate in recent years. An increasing number of vehicles are equipped with different levels of driving assist systems to reduce the human driver’s burden. However, because of the conservative design of its programming framework, there is still a large gap between the performance of current autonomous driving systems and experienced veteran drivers. This gap can cause drivers to distrust decisions or behaviors made by autonomous vehicles, thus affecting the effectiveness of drivers’ use of auto-driving systems. To further estimate the expected acceptance of autonomous driving systems in real human–machine co-driving situations, a characterization model of driver confidence has to be constructed. This paper conducts a survey of driver confidence in riding autonomous vehicles. Based on the analysis of results, the paper proposes a confidence quantification model called “the Virtual Confidence (VC)” by quantifying three main factors affecting driver confidence in autonomous vehicles, including (1) the intrusive movements of surrounding traffic participants, (2) the abnormal behavior of the ego vehicle, and (3) the complexity of the driving environment. The model culminates in a dynamic confidence bar with values ranging from 0 to 100 to represent the levels of confidence. The validation of the confidence model was verified by doing comparisons between the real-time output of the VC and the real-time feeling of human drivers on an autonomous vehicle simulator. The proposed VC model can potentially identify features that need improvement for auto-driving systems in unmanned tests and provide data reference.

Autonomy and mobility simulation time reduction through machine learning while considering uncertainty and reliability prediction

Modeling and simulation (M&S) tools are used extensively throughout the Ground Vehicle Systems Center and the U.S. Army to perform an analysis of ground vehicles more quickly and less expensively than through physical testing. The Computational Research and Engineering Acquisition Tools and Environments-Ground Vehicles (CREATE-GV) project is an M&S software effort that focuses on mobility and autonomous vehicle simulation and analysis, using physics-based three-dimensional modeling to accurately calculate a variety of ground vehicle metrics and parameters of interest. However, because these simulations are high fidelity, they often require a great deal of computational power and time. One approach to reducing simulation time that has proved effective in certain contexts is the creation of “surrogate models” through machine learning (ML) algorithms. However, it is often very challenging to accurately predict the mobility of a ground vehicle system in general, and there is no existing model that can predict the mobility of autonomous systems. A great deal of uncertainty exists in the mobility and autonomy area of physics-based simulation models related to modeling assumptions, terrain conditions, and insufficient knowledge of interactions between the vehicle and terrain. Understanding how the uncertainties inherent in autonomous mobility prediction affect model accuracy is still an open fundamental research question. We present a surrogate modeling approach leveraging ML algorithms to work with CREATE-GV to increase the computation speed of mobility assessments while still considering the reliability of the mobility predictions under uncertainty.

Effects of Sim2Real Image Translation via DCLGAN on Lane Keeping Assist System in CARLA Simulator

Autonomous vehicle (AV) simulation using a virtual environment has the advantage of being able to test algorithms in various scenarios with reduced resources. However, there may exist a visual gap between the virtual environment and the real-world. In this paper, in order to mitigate this gap, we trained Dual Contrastive Learning Generative Adversarial Networks (DCLGAN) to realistically convert the image of the CARLA simulator and then evaluated the effect of the Sim2Real conversion focusing on the lane keeping assist system (LKAS). Moreover, in order to avoid the case where the lane is translated distortedly by DCLGAN, we found the optimal training hyperparameters using feature similarity (FSIM). After training, we built a system that connected the CARLA simulator with DCLGAN and AV in real-time. As for the result, we collected data and analyzed them using the following four methods. First, image reality was measured with Fréchet Inception Distance (FID), which we quantitatively verified to reflect the lane characteristics. The CARLA images that passed through DCLGAN had smaller FID values than the original images. Second, lane segmentation accuracy through ENet-SAD was improved by DCLGAN. Third, in the curved route, the case of using DCLGAN drove closer to the center of the lane and had a high success rate. Lastly, in the straight route, DCLGAN improved lane restoring ability after deviating from the center of the lane as much as in reality. Consequently, it convinced that the proposed method could be applicable to mitigate the gap of simulation toward real-world.

Developing Requirements to Validate Autonomous Ground Vehicle Simulations

This project developed a methodology for assessing simulation testing of the United States’ military autonomous vehicle platforms. The authors conducted background research of autonomous vehicles and simulation to determine the best virtual testing methods that provided the strongest evidence of simulation performance. The team created a requirements list for the simulation software of autonomous vehicles to help drive the virtual test development and conducted statistical analyses, pairwise comparisons, and visual analysis tests to assess simulation data compared to data from physical test runs. The authors identified multiple methods of virtual testing that can assess the autonomous vehicle simulation for further development.

Training Artificial Intelligence Algorithms with Automatically Labelled UAV Data from Physics-Based Simulation Software

Machine-learning (ML) requires human-labeled “truth” data to train and test. Acquiring and labeling this data can often be the most time-consuming and expensive part of developing trained models of convolutional neural networks (CNN). In this work, we show that an automated workflow using automatically labeled synthetic data can be used to drastically reduce the time and effort required to train a machine learning algorithm for detecting buildings in aerial imagery acquired with low-flying unmanned aerial vehicles. The MSU Autonomous Vehicle Simulator (MAVS) was used in this work, and the process for integrating MAVS into an automated workflow is presented in this work, along with results for building detection with real and simulated images.

Effective AV Simulation

Simphera from dSPACE is bringing speed and simplicity to the simulation and validation of autonomous vehicles

Recreating Reality

Claytex is adopting a holistic approach to the simulation of autonomous vehicles, explains Theodor Ensbury, one of the company's simulation engineers

Lane-Changing Trajectory Tracking and Simulation of Autonomous Vehicles Based on Model Predictive Control

In order to realize the lane-changing maneuver of Connected Autonomous Vehicles (CAV), a lateral controller based on model predictive control is developed with the three degrees of freedom vehicle dynamic model. Then the controller is synthesized to track the reference trajectory fitted by the quintic Bézier curve. The controller is validated by MATLAB/CarSim under different road adhesion conditions and driving speeds. Results show that for different road adhesion conditions and driving speeds, the controller does not need to adjust the control parameters and can continuously correct the deviation from the expected trajectory. During the tracking process, the front wheel angle, front wheel angle increment, centroid side deflection angle, and front wheel side deflection angle are kept within the limited constraint range. The established control algorithm has good control robustness and tracking driving stability. The research can provide a theoretical basis and technical support for lane-changing safety and control of CAV.

The Simulation of Autonomous Vehicle Using ROS2 Based on Convolutional Neural Networks for Object Recognition

The main justification for implementing an Autonomous Vehicle (AV) system in the real world is the safety aspect of driving, because if there is an error in driving then the error will become a gap that can threaten the safety of the driver himself and other drivers, therefore an AV system is made to reduce driver errors. in driving. The aim of this research is to implement one of the parts of the AV system, that is object recognition, and in this study, we also conduct an experiment with simulating the object recognition feature that has been implemented in order to get more concrete results. Architectural object recognition is designed to extract key features from traffic sign images, the traffic sign detection uses the customized Convolutional Neural Networks (CNNs) architecture. After the architectural has been implemented, training will be carried out using Custom Traffic Sign Dataset and experiments will also be conducted to simulate object recognition by applying ROS2 as a car robotic system that represents a car's functionality system in the real world. the results of this study for the implementation of the modified CNNs architecture is 99.96% and the results of the simulations carried out show that the prototype can detect traffic signs objects with a distance of 10m

Simulation and control of a novel hovering type autonomous underwater vehicle (HAUV) with ballast tanks

Simulation and control of a novel hovering type autonomous underwater vehicle (HAUV) with ballast tanks

Human‐Like Interactive Behavior Generation for Autonomous Vehicles: A Bayesian Game‐Theoretic Approach with Turing Test

Interacting with surrounding road users is a key feature of autonomous vehicles and is critical for their intelligence testing. Existing interaction modalities in autonomous vehicle simulation and testing are not sufficiently smart, and they barely reflect human‐like behaviors in real‐world driving scenarios. Therefore, a novel hierarchical game‐theoretical framework is presented to represent naturalistic multimodal interactions among road users during simulation and testing, which is subsequently validated by the Turing test. Given that human drivers have no access to the complete information of the surrounding road users, the Bayesian game theory is used to model the decision‐making process. Next, a probing behavior is generated by the proposed game‐theoretic model, and is further applied to control the vehicle via the Markov chain. The proposed method is tested through a series of experiments and compared with existing approaches. In addition, Turing tests are conducted to quantify the human‐likeness of the proposed algorithm. The experimental results show that the proposed Bayesian game‐theoretic framework can effectively generate representative scenes of human‐like decision‐making during autonomous vehicle interactions, demonstrating its feasibility and effectiveness. An interactive preprint version of the article can be found at: https://www.authorea.com/doi/full/10.22541/au.163482630.00283278.