Ground Vehicle Simulation Research Articles

Geometric Fidelity Requirements for Meshes in Automotive Lidar Simulation

The perception of vegetation is a critical aspect of off-road autonomous navigation, and consequentially a critical aspect of the simulation of autonomous ground vehicles (AGVs). Representing vegetation with triangular meshes requires detailed geometric modeling that captures the intricacies of small branches and leaves. In this work, we propose to answer the question, “What degree of geometric fidelity is required to realistically simulate lidar in AGV simulations?” To answer this question, in this work we present an analysis that determines the required geometric fidelity of digital scenes and assets used in the simulation of AGVs. Focusing on vegetation, we use a comparison of the real and simulated perceived distribution of leaf orientation angles in lidar point clouds to determine the number of triangles required to reliably reproduce realistic results. By comparing real lidar scans of vegetation to simulated lidar scans of vegetation with a variety of geometric fidelities, we find that digital tree models (meshes) need to have a minimum triangle density of >1600 triangles per cubic meter in order to accurately reproduce the geometric properties of lidar scans of real vegetation, with a recommended triangle density of 11,000 triangles per cubic meter for best performance. Furthermore, by comparing these experiments to past work investigating the same question for cameras, we develop a general “rule-of-thumb” for vegetation mesh fidelity in AGV sensor simulation.

Interval Type 2 Fuzzy Adaptive Motion Drive Algorithm Design

Motion drive algorithms are a set of filters designed to simulate realistic motion and are an integral part of contemporary vehicle simulators. This paper presents the design of a novel intelligent interval type 2 fuzzy adaptive motion drive algorithm for an off-road uphill vehicle simulator. The off-road, uphill vehicle simulator is used to train and assess the driver’s behavior under varying operational and environmental conditions in mountainous terrain. The proposed algorithm is the first of its kind to be proposed for off-road uphill vehicle simulators, and it offers numerous benefits over other motion drive algorithms. The proposed algorithm enables the simulator to adapt to changes in the uphill road surface, vehicle weight distribution, and other factors that influence off-road driving in mountainous terrain. The proposed algorithm simulates driving on hilly terrain more realistically than existing algorithms, allowing drivers to learn and practice in a safe and controlled environment. Additionally, the proposed algorithm overcomes limitations present in existing algorithms. The performance of the proposed algorithm is evaluated via test drives and compared to the performance of the conventional motion drive algorithm. The results demonstrate that the proposed algorithm is more effective than the conventional motion drive algorithm for the ground vehicle simulator. The pitch and roll responses demonstrate that the proposed algorithm has enabled the driver to experience abrupt changes in terrain while maintaining the driver’s safety. The surge response demonstrated that the proposed MDA handled the acceleration and deceleration of the vehicle very effectively. In addition, the results demonstrated that the proposed algorithm resulted in a smoother drive, prevented false motion cues, and offered a more immersive and realistic driving experience.

Vibration quality and ride comfort investigation with transient excitation in a ground vehicle simulator environment

Ground vehicle ride comfort is closely associated with vibration quality as perceived by the human occupants of a vehicle. While outstanding vibration quality is a desirable outcome of the vehicle design process, it is challenging to relate customer perceptions of vibration quality to customer-facing vehicle responses measured inside of a vehicle. Previous work with steady-state vibration has shown that human perception of vibration can be compared to how humans perceive sound, where both the amplitude and frequency of the excitation will alter its perception. Thus, a simple measurement of vibration level is insufficient to predict comfort. In this work, an analysis of human perception of transient vibration is undertaken using a ground vehicle simulator. Physical measurements from inside four different production vehicles are presented including accelerometer, microphone, and action camera recordings. The test event performed is travel over a cleat, with varying vehicle speeds and tire inflation pressures. The physical measurements are used as inputs to the simulator. A comparison is made between the perceived vibration quality of the vehicles as reported by an expert human jury and the physically measured vehicle responses. A ranking of 36 cleat test events for vibratory harshness from best to worst is presented.

AutoVRL: A High Fidelity Autonomous Ground Vehicle Simulator for Sim-to-Real Deep Reinforcement Learning

Deep Reinforcement Learning (DRL) enables cognitive Autonomous Ground Vehicle (AGV) navigation utilizing raw sensor data without a-priori maps or GPS, which is a necessity in hazardous, information poor environments such as regions where natural disasters occur, and extraterrestrial planets. The substantial training time required to learn an optimal DRL policy, which can be days or weeks for complex tasks, is a major hurdle to real-world implementation in AGV applications. Training entails repeated collisions with the surrounding environment over an extended time period, dependent on the complexity of the task, to reinforce positive exploratory, application specific behavior that is expensive, and time consuming in the real-world. Effectively bridging the simulation to real-world gap is a requisite for successful implementation of DRL in complex AGV applications, enabling learning of cost-effective policies. We present AutoVRL, an open-source high fidelity simulator built upon the Bullet physics engine utilizing OpenAI Gym and Stable Baselines3 in PyTorch to train AGV DRL agents for sim-to-real policy transfer. AutoVRL is equipped with sensor implementations of GPS, IMU, LiDAR and camera, actuators for AGV control, and realistic environments, with extensibility for new environments and AGV models. The simulator provides access to state-of-the-art DRL algorithms, utilizing a python interface for simple algorithm and environment customization, and simulation execution.

Learning multiobjective rough terrain traversability

We present a method that uses high-resolution topography data of rough terrain, and ground vehicle simulation, to predict traversability. Traversability is expressed as three independent measures: the ability to traverse the terrain at a target speed, energy consumption, and acceleration. The measures are continuous and reflect different objectives for planning that go beyond binary classification. A deep neural network is trained to predict the traversability measures from the local heightmap and target speed. To produce training data, we use an articulated vehicle with wheeled bogie suspensions and procedurally generated terrains. We evaluate the model on laser-scanned forest terrains, previously unseen by the model. The model predicts traversability with an accuracy of 90%. Predictions rely on features from the high-dimensional terrain data that surpass local roughness and slope relative to the heading. Correlations show that the three traversability measures are complementary to each other. With an inference speed 3000 times faster than the ground truth simulation and trivially parallelizable, the model is well suited for traversability analysis and optimal path planning over large areas.

Local Path-Planning Simulation and Driving Test of Electric Unmanned Ground Vehicles for Cooperative Mission with Unmanned Aerial Vehicles

Recently, various studies related to the development of unmanned vehicles have been conducted around the world. In particular, unmanned ground vehicles (UGVs) and unmanned aerial vehicles (UAVs) have been developed and utilized for various purposes. In this study, we developed a method for the path generation of UGVs in a system in which one operator controls many different types of unmanned vehicles. In the driving control system (DCS), it is necessary to process sensor data such as GPS/INS and LiDAR when generating a path by receiving the target waypoint from the ground control station. In addition, the DCS must upload the current location, posture, state, etc., as well as save driving log. Therefore, in order to recognize obstacles in real time and generate a path, a safe path generation algorithm with a short computation time is required. Among the various path generation methods, the potential field algorithm was selected, and the algorithm was modified to reduce the computation time. The computation time before and after modification of the algorithm was obtained and compared through simulation, and the algorithm was verified through application to an actual system by performing an obstacle avoidance experiment and a simultaneous control experiment for two UGVs.

Chrono::Vehicle: template-based ground vehicle modelling and simulation

Chrono::Vehicle: template-based ground vehicle modelling and simulation

Chrono::Vehicle: template-based ground vehicle modelling and simulation

Chrono::Vehicle is a module of the open-source multi-physics simulation package Chrono, aimed at modelling, simulation, and visualisation of wheeled and tracked ground vehicle multi-body systems. Its software architecture and design was dictated by the desire to provide an expeditious and user friendly mechanism for assembling complex vehicle models, while leveraging the underlying Chrono modelling and simulation capabilities, allowing seamless interfacing to other optional Chrono modules (e.g., its granular dynamics and fluid-solid interaction (FSI) capabilities), and providing a modular and expressive API to facilitate its use in third-party applications. Vehicle models are specified as a hierarchy of subsystems, each of which is an instantiation of a predefined subsystem template. Written in C++, Chrono::Vehicle is offered as a middleware library. In this paper, we provide an overview of the Chrono::Vehicle software design philosophy, its main capabilities and features, describe the types of ground vehicle mobility simulations it enables, and outline several directions of future development and planned extensions.

실시간 주행성 분석 알고리즘 검증을 위한 무인 자율 주행 차량 가상 시뮬레이터

실시간 주행성 분석 알고리즘 검증을 위한 무인 자율 주행 차량 가상 시뮬레이터

공학 컴포넌트 모델을 가지는 시뮬레이터 연동구조 설계

This paper include the results of the software interface design for the ground vehicle simulators. This design allows us to upgrade the simulator easily by exchanging the engineering model when we need to change the function of algorithm or data of a simulator. That is because a main part of simulator is composed of reusable model.

Advanced structure methodologies for next-generation ground vehicles. Part 2: Case study

In the first paper of this two-part study, basic theories were introduced for several new methodologies developed in the Automotive Research Center (a US Army TACOM Center of Excellence for Modelling and Simulation of Ground Vehicles at the University of Michigan) for the simulation and design of advanced structures and materials for next-generation ground vehicles. These new methodologies include: (1) an advanced topology optimisation technique for innovative conceptual design of vehicle structures and materials; (2) a systematic design optimisation process with efficient analysis and sensitivity analysis capabilities for detailed design modifications to improve the vibration and noise characteristics of a complex vehicle structure; (3) a reduced-order modelling technique that can be used to systematically generate low-order models for the prediction of vehicle vibration, power flow, and the effects of parameter uncertainties; and (4) an efficient and accurate energy boundary element analysis method for high-frequency noise analysis outside the vehicle. In this second paper, an extensive case study is presented to demonstrate how the methodologies presented in the first paper can be applied to a vehicle system. A pick-up truck equipped with an advanced hybrid propulsion system is considered in this paper, and various example design and prediction problems are discussed, which provide proof-of-concept for the methodologies developed.

Advanced structure methodologies for next-generation ground vehicles. Part 1: Basic theories

One of the major objectives of the Automotive Research Center (a US Army TACOM Center of Excellence for the modelling and simulation of ground vehicles at the University of Michigan) is to develop new methodologies for advanced structures and materials for next-generation ground vehicles. Several major developments in this area are detailed in this paper. First, an advanced topology optimisation technique is presented, which provides a tool for laying out new, conceptually advanced designs for vehicle structures, or substructures, to achieve the lower weight and higher performance requirements for next-generation ground vehicles. Second, a 'sizing' design optimisation process is presented for detailed design changes in order to improve the vibro-acoustic response of a complex vehicle structure. This process incorporates efficient analysis and sensitivity analysis capabilities for vibro-acoustic systems. In addition, a component-based technique is presented for generating reduced-order models of a vehicle structure in order to lower the computational costs of vibration analysis. This technique is also extended to analysing vibration transmission in a complex vehicle structural system to determine the power flow among components and the effect of parameter uncertainties. Finally, an energy boundary element analysis method is presented for efficient and accurate high-frequency noise analysis, which extends the capability for predicting the acoustic field around the vehicle due to various sources.

Reusing a Simulation Facility Software Architecture for Embedded Simulation

The authors take a ground vehicle simulation software architecture, which was designed for stationary simulation in crewstations in a laboratory environment, and reuse it to add embedded simulation capabilities to fielded ground vehicles. The purpose of embedded simulation is to enhance soldier effectiveness both before and during actual battlefield conditions. As a result, in a fielded ground vehicle, the computer specialist responsible for setting up, starting, controlling, monitoring, and stopping the simulation is not present. Unlike in a laboratory environment, in a fielded vehicle, the simulation software itself is responsible for performing these functions. The authors provide a case study illustrating characteristics of a system architecture that allow it to be long-lived by evolving not only to changing requirements due to technological and methodological changes but also to changes due to new and previously unthought-of patterns of system utilization.

TRANSPORTATION SAFETY RESEARCH APPLICATIONS UTILIZING HIGH-FIDELITY DRIVING SIMULATION

The U.S. Department of Transportation's National Highway Traffic Safety Administration constructed the National Advanced Driving Simulator (NADS), the highest fidelity ground vehicle simulator in the world, to investigate human-centered issues as they relate to driving safety. The simulator's primary mission is to study causes of traffic collisions, with the aim of reducing fatalities on U.S. highways. This paper provides an overview of the design, core capabilities, and key applications of the NADS and discusses how they may influence safety.

Ground vehicle modeling and simulation of military vehicles using high performance computing

In support of the Department of Defense (DOD) mandated acquisition reform initiatives to reduce vehicle-related life cycle costs and time-lines, the Tank-automotive and Armaments Research, Development and Engineering Center (TARDEC) is using simulation-based acquisition strategies to investigate the dynamic performance of wheeled and tracked ground vehicles throughout the vehicle development, testing, and fielding life cycle process. The paper will describe how modeling and simulation (M&S) is applied to answer a wide variety of design and performance evaluation questions and will depict a series of simulation-based engineering projects that build on the Army's simulation investments as a tool to investigate and answer real-world vehicle design, acquisition, and engineering support questions.

Characterizing Shock Absorbers for Ground Vehicle Simulation

Abstract A new method of ground vehicle shock absorber characterization using lumped acceleration excitation is presented. By using a triangular waveform and lumping the acceleration at the beginning and end of travel, this method produces a 27% larger mapping on the phase plane than traditional harmonic (sinusoidal waveform) excitation characterization. Force response surfaces are created based on the lumped acceleration and on harmonic excitation data. A difference surface is created between the common regions of both force response surfaces. A correlation is made between the response of the lumped acceleration force response surface to harmonic inputs and observed data. Hysteresis effects are noted using the lumped acceleration force response surface at low frequencies. A standard materials testing machine is used to characterize the shock absorber.

5.4.2 Mousetrap: A Concurrent Engineering Simulation That Keeps Going…and going…and going

AbstractThe Department of Defense Systems Management College uses the Stored Energy Ground Vehicle Simulation (“Mousetrap”) to train future program managers in the application and power of concurrent engineering and the systems engineering process. The simulation has evolved over the past seven years into what we believe to be a world class concurrent engineering simulation, which additionally provides the military officers and DoD civilians a critically important industry perspective.Each student integrated product team performs the role of a defense contractor competing against other student “companies” in their quest for the single follow‐on production contract to be awarded after completion of the acquisition development cycle and a government supervised competitive “run‐off”. They respond to a government request for proposal, negotiate a contract and spend time over 14 weeks developing a life cycle balanced system, expending and accounting for labor and actual government funds for materials. They must conduct three acquisition phase based technical design reviews presented to an integrated government program office, and subsequently build and field a competitive prototype.

The Iowa Driving Simulator: a tool for human factors and vehicle virtual prototyping research

The Iowa Driving Simulator (IDS) is a six degree–of–freedom, hexapod configuration, ground vehicle simulator located at the Center for Computer–Aided Design (CCAD) at The University of Iowa. The simulator is designed for use in human factors and vehicle virtual prototyping research studies. Currently one of the most advanced ground vehicle simulators in the world, the IDS relies on high–fidelity simulation software and hardware. In the context of this work, high–fidelity is meant to imply: high–resolution, photo textured graphics, real–time multibody vehicle dynamics simulation, and appropriate sound, control loading and motion cues. The IDS incorporates all of these features in a flexible system capable of providing repeatable human factors experiments at a level of simulation detail which is useful for engineering design.

Real-Time Simulation of Multibody Dynamics on Shared Memory Multiprocessors

This paper presents a recursive variational formulation for real-time simulation of multibody mechanical systems on shared memory parallel computers. Static scheduling algorithms are employed to evenly distribute computation on shared memory multi-processors. Based on the methods developed, a general-purpose dynamic simulation program is shown to simulate multibody systems faster than real-time, enabling operator-in-the-loop simulation of ground vehicles and robots.