Browse Topic: Software-in-the-loop (SIL)

Items (134)
This paper proposes a nonlinear and robust State-Dependent Riccati Equation (SDRE) combined with H∞ control architecture for brake- by-wire systems, specifically designed to handle severe tire-road friction variations and μ-split scenarios. The primary objective is to maximize deceleration capabilities while rigorously maintaining yaw stability, trajectory tracking, and passenger comfort through jerk limitation. Situated within the domain of active safety, this research addresses robustness against real-world uncertainties by utilizing a high-fidelity 14-degree-of-freedom vehicle model that accounts for longitudinal, lateral, and yaw dynamics, suspension-induced pitch and roll effects, and nonlinear tire behavior with explicit load transfer. To ensure near-optimal slip tracking under variable surface conditions, the system employs online friction estimation via Extended and Unscented Kalman Filters (EKF/UKF) fusing wheel and IMU data to adaptively adjust slip targets. The control strategy is bifurcated: the SDRE component manages dominant nonlinearities through state-dependent gains to prevent wheel lock-up, while the H∞ component provides robust disturbance rejection against parametric uncertainties such as mass variations and sensor noise. Control efforts are distributed via a Quadratic Programming (QP) torque allocator featuring anti-windup mechanisms and explicit saturation handling to compensate for lateral drift during μ-split braking. Validation is conducted through a Model- in-the-Loop (MIL) to Software-in-the-Loop (SIL) pipeline using scenarios including wet surfaces and panic braking. Simulation results demonstrate enhanced yaw stability and controlled deceleration profiles compared to conventional baselines, ensuring computational feasibility for automotive Electronic Control Units (ECUs).
Cubillos, Ximena Celia Méndez
Building upon previous work that successfully employed a Reinforcement Learning (RL) agent for the autonomous optimization of transmission shift programs to enhance fuel efficiency, this paper addresses a critical limitation of that approach: the neglect of human-centric factors. While the prior methodology achieved substantial fuel consumption reductions by training an RL agent in a Software-in-the-Loop (SiL) environment, it did not explicitly account for aspects such as driver comfort and preferences, which are paramount for real-world user acceptance and drivability. This work presents a multi-objective optimization framework extending the artificial calibrator to simultaneously maximize fuel efficiency and enhance driver comfort. The method introduces a modified RL reward function that penalizes undesirable shift behavior to ensure a smooth driving experience (drivability). This new methodology also incorporates a mechanism to capture and integrate driver preferences, moving beyond a purely quantitative fuel-economy-driven objective to a holistic, user-focused calibration. Experimental evaluation demonstrates that the extended framework successfully generates a shift strategy that achieves a favorable trade-off between fuel efficiency and drivability, resulting in a more balanced and practical calibration. The ability to integrate these qualitative factors into an automated, data-driven process represents a significant step forward, promising to accelerate the development of powertrain control systems that are both highly efficient and aligned with the expectations of human drivers. This work lays the foundation for future RL-based calibration tools that are capable of addressing the full spectrum of development objectives, from fuel economy to the subtleties of vehicle drivability.
Kengne Dzegou, Thierry JuniorSchober, FlorianRebesberger, RonHenze, RomanSturm, Axel
With the increasing market penetration of automated vehicles, there is a critical need for credible and repeatable methods to quantify their energy impacts. This paper presents a Model-Based Systems Engineering (MBSE)-driven Anything-in-the-Loop (XIL) methodology for quantifying the powertrain energy consumption and potential savings from various controls for automated vehicles in realistic road scenarios while preserving high-fidelity powertrain behavior. The novelty of this approach lies in its use of a unified MBSE backbone (AMBER: Argonne National Laboratory’s [Argonne’s] MBSE-centric platform for transportation energy analysis) to automate the seamless and traceable progression from pure simulation to Vehicle-in-the-Loop (VIL) testing. This work utilizes Argonne's multi-vehicle simulation tool, RoadRunner, which automatically constructs closed-loop road scenarios (road geometry, vehicle sensors, other vehicles, and traffic controls) and connects them to Argonne’s validated, high-fidelity vehicle and powertrain models in Autonomie. The MBSE backbone in AMBER organizes requirements, interfaces, plant and controller models, and test scenarios into a single set of models that is maintained across pure simulation, Software-in-the-Loop (SIL), Processor-in-the-Loop (PIL), and VIL stages. Each stage has a clear role: simulation enables rapid development and validation of advanced models or controls across a large number of scenarios; SIL supports standalone algorithm verification and scenario down-selection; PIL validates real-time execution, inputs/outputs, and timing on the target processor; and VIL provides closed-loop evaluation with a real vehicle under controlled laboratory conditions. AMBER’s automated build and configuration enable rapid retargeting across platforms and repeatable scenario reproduction, making validation fast and cost-effective. To demonstrate its practical application, the workflow is used to validate the functionality and quantify the energy savings of an eco-driving control against a calibrated human driver model. Experiments show strong repeatability and consistent energy gains for the eco-driving strategy while preserving trip time, yielding average energy savings of 7.8% across the evaluated scenarios. Overall, the MBSE-guided XIL workflow shortens development time and reduces test cost by limiting on-road testing and lowering integration risk before track evaluation, while producing credible, closed-loop energy assessments traceable from requirements to test evidence.
Jeong, JongryeolSharer, PhillipDi Russo, MiriamDas, DebashisZhang, YaozhongKarbowski, Dominik
In the automotive industry, increasing noise regulations are influencing product sales and passenger comfort, creating a need for more effective noise testing methods. Hardware-in-Loop (HiL) based virtual acoustic testing serves as a critical step before Driver-in-Loop testing, allowing for the assessment of vehicle performance and noise levels inside and outside the vehicle under various conditions before physical prototype testing is performed. The Hardware-in-the-Loop (HiL) simulator setup is equipped with joystick control that requires a physical representation of the vehicle dynamics model provided as a Functional Mock-up Unit (FMU) in real-time format. In contrast, the vehicle control logic is implemented in C++ code. The simulator incorporates both lateral and longitudinal dynamics. Additional interfaces are integrated to support joystick input and virtual road visualization enabling realistic vehicle maneuvering and dynamic performance evaluation. However, performing all test protocols directly on the HiL setup can be time-consuming and costly. To address this limitation of full HiL testing, in this study, an offline Software-in-the-Loop (SiL) Co-simulation framework was developed as an alternative. This method replicates the HiL environment within MATLAB/Simulink, where joystick actions are simulated according to predefined driving protocols. The dynamic behavior of the vehicle during a reverse driving protocol, involving a 540° constant steering angle and 0–100% acceleration pedal input, was analyzed and compared between Offline SiL and HiL environments. Results demonstrated that 85% of key parameters exhibited strong correlation (R2 > 0.9), confirming that the offline SiL-based approach effectively replicates HiL performance. The remaining parameters also showed acceptable consistency. These findings indicate that the proposed Offline Co-simulation method is a promising, cost-effective, and scalable alternative for accurately predicting vehicle dynamic behavior, aligning well with current automotive industry needs for early-stage validation and optimization.
Visuvamithiran, RishikesanChougule, SourabhSrinivasan, RangarajanLaurent, Nicolas
Advancing HIL and SIL Validation for eVTOL from Tip to Battery to Tail by Bloomy Controls
Blume, Peter
Ensuring the safety and functionality of sophisticated vehicle technologies has grown more difficult as the automotive industry quickly shifts to intelligent, electric, and connected mobility. Software-defined architectures, electric powertrains, and advanced driver assistance systems (ADAS) all require strong quality assurance (QA) frameworks that can handle the multi domain nature of contemporary vehicle platforms. In order to thoroughly assess the functionality and dependability of next generation automotive systems, this paper proposes an integrated QA methodology that blends conventional testing procedures with model-based validation, digital twin environments, and real-time system monitoring. The suggested framework, which includes hardware-in-the-loop (HIL), software-in-the-loop (SIL), and over-the-air (OTA) testing techniques, concentrates on end-to-end traceability from specifications to validation. Simulating intricate situations for ADAS, electric vehicle battery temperature management, and dynamic system updates in connected platforms are prioritized. This study also outlines the main obstacles to integrating QA methods with changing regulatory environments and draws attention to discrepancies between operational performance in real-world scenarios and compliance benchmarks. Early fault detection, lifecycle validation, and continuous improvement are made possible by the QA process's transition from reactive to proactive through the integration of digital twins and predictive analytics. A strategic roadmap for QA specialists and test engineers to adjust to changing industry demands is presented in the paper's conclusion. In addition to promoting safety and dependability, the suggested framework speeds up time to market, lowers development costs, and increases consumer confidence in cutting-edge automotive technologies.
Komanduri, Arun SrinivasSrivastava, Anuj
Functional Mock-up Units (FMUs) have become a standard for enabling co-simulation and model exchange in vehicle development. However, traditional FMUs derived from physics-based models can be computationally intensive, especially in scenarios requiring real-time performance. This paper presents a Python-based approach for developing a Neural Network (NN) based FMU using deep learning techniques, aimed at accelerating vehicle simulation while ensuring high fidelity. The neural network was trained on vehicle simulation data and trained using Python frameworks such as TensorFlow. The trained model was then exported into FMU, enabling seamless integration with FMI-compliant platforms. The NN FMU replicates the thermal behavior of a vehicle with high accuracy while offering a significant reduction in computational load. Benchmark comparisons with a physical thermal model demonstrate that the proposed solution provides both efficiency and reliability across various driving conditions. The paper discusses the workflow for model training and integration strategies for deep learning models within simulation tools like AMESIM and Simulink. With this approach significant time reduction is observed without affecting the accuracy when compared with the physical model. NN FMU also reduces efforts up to 40 % compared with traditional FMU conversion. CPU improvement from physical to NN FMU model achieved greater than 30 % reduction with the same accuracy. NN FMU maintains FMI compatibility and can be directly used in a wide range of XiL applications such as Model-in-the-Loop (MiL), Software-in-Loop (SiL), and Hardware-in-Loop (HiL) testing scenarios. This NN FMUs opens pathways for hybrid modelling approaches that combine data-driven and physics-based paradigms for automotive simulations.
Srinivasan, RangarajanAshok Bharde, PoojaMhetras, MayurChehire, Marc
In the current automotive design and development of the Electrical Distribution System (EDS), at an earlier stage, before the physical prototyping is largely absent. Traditional methods for verification and validation of EDS are performed with HIL, SIL, MIL, prototype testing or physical vehicle trials reveal design errors at later stages in the development cycle, which may lead to redesign, prolonged timelines and increased failure rates at vehicle integration. Hence, there is a critical need for an early-stage simulation methodology that ensures robustness and reliability of E/E architecture with first-time-right readiness at the design stage itself. In this paper, a digital EDS architecture simulation introduces a mode-based structural behavioural approach where specific vehicle functions, failure conditions and malfunction scenarios are set up in a simulation environment with their corresponding electrical circuits for simulation. A function-specific truth table-based analysis model enabling the controller to control the electrical paths for different electrical loads dynamically. This methodology ensures digital verification of electrical loads behaviour at different operating conditions, power distribution and switching logics are accurately validated during the design stage, reducing production time issues and ensuring seamless transition to series production.
Jaisankar, GokulnathWarke, UmakantChakra, PipunBorole, Akash
Currently, we face the challenge that ensuring ADS safety remains the primary bottleneck to large-scale commercial deployment—while benchmarks such as the CARLA Leaderboard have spurred progress, their coarse evaluation granularity, inability to quantify procedural risks, and lack of differentiation among algorithms in complex scenarios make in-depth diagnostics and functional safety validation exceedingly difficult. To address these challenges, we propose EvalDrive, a framework that seems to offer a more comprehensive approach to multi-scenario performance evaluation for modular autonomous driving systems. Within this broader analytical framework, EvalDrive appears to provide what seems to be three key contributions. (1) It constructs what appears to represent a structured and extensible scenario library, comprising a majority of 44 interactive scenarios, 23 weather conditions, and 12 town environments, which are then systematically expanded through parameterized variations. (2) Our paper presents a multi-dimensional evaluation approach that shifts the emphasis from outcome-based safety to process-oriented safety, enabling the quantification of near-collision behaviors. Moreover, context-aware metrics—such as the Index of Driving Efficiency (IDE)—are employed to characterize distinct driving styles. (3) The framework further implements a highly integrated closed-loop co-simulation platform. By tightly coupling the CARLA simulator with the Apollo ADS, it establishes a high-fidelity, reproducible software-in-the-loop (SIL) environment. What this pattern seems to suggest, therefore, is that EvalDrive provides what appears to be a more comprehensive paradigm—from scenario construction and multi-dimensional evaluation to closed-loop validation—seeming to offer more robust diagnostic support for the iterative optimization and safe deployment of autonomous driving systems.
Jia, ChunyuKong, YanMa, YaoPei, Xiaofei
This paper presents a comprehensive analysis of advanced methods for optimizing software development in hybrid vehicles, focusing on the V-Model methodology integrated with Model-Based Systems Engineering (MBSE), functional design techniques and In-the-Loop validation processes, and the incorporation of agile methodologies such as SAFe (Scaled Agile Framework). The increasing complexity of embedded systems in hybrid vehicles, driven by electrification and the introduction of autonomous and connected systems, demands systematic and rigorous approaches to ensure reliability, safety, and energy efficiency. Over the next sections, we will explore the fundamental principles of the V-Model, its adaptations to the context of hybrid vehicles, the implementation of functional design processes supported by MBSE, the application of Software-in-the-Loop (SiL) and Hardware-in-the-Loop (HiL) methodologies for system validation, and finally the integration of agile SAFe principles to manage complexity at scale. The goal is to provide a detailed technical overview that helps engineers and researchers implement more efficient development processes, reducing time to market while maintaining the robustness required for critical automotive systems.
Gomes, Cleber WillianNatal, Icarus Lima
This paper offers recent ideas and its implementation on leveraging AI for off highway Autonomous vehicle Simulations in SIL and HIL frameworks. Our objective is to enhance software quality and reliability while reducing costs and efforts through advanced simulation techniques. We employed multiple innovative solutions to build a System of Systems Simulation. Physics based models are a prerequisite for detailed and accurate representation of the real-world system, but it poses challenges due to its computational complexity and storage requirements. Machine learning algorithms were used to create surrogate/reduced order models to optimize by preserving the expected fidelity of models. It helped to speed up simulation and compile model code for SIL & HIL Targets. Built AI driven interfaces to bridge windows, Linux and Mobile Operating systems. Time synchronization was the key challenge as multiple environments were needed for end-to-end solutions. This was resolved by reinforcement learning & optimization algorithms so that loss of information can be prevented. John Deere Operations Center™ Fleet management was integrated with vehicle simulators so that remote monitoring and control of machine configurations and settings for various autonomy mode could be validated in virtual environment. Gen AI based tools were used for creation of Test plan and its Automation to accumulate several hundred hours of test execution for autonomy related features. The team tested various SW & HW fault conditions to understand the impact and behavior of the system in autonomy mode. As part of the next steps this framework would be further scaled for future autonomy programs and product lines This adoption of AI-based methods has expedited the delivery of autonomous vehicles, ensuring they are technologically advanced and customized to meet customer needs.
Karegaonkar, Rohit P.Aole, SumitDasnurkar, SwapnilSingh, VishwajeetSaha, Soumyadeep
Evaluating the impact of software changes on fuel consumption and emissions is a critical aspect of transmission development. To evaluate the trade-offs between performance improvements and potential negative effects on efficiency, a forward-looking Software-in-the-Loop (SiL) simulation has been developed. Unlike backward calculations that derive fuel consumption based solely on cycle speed and engine speed, this approach executes complete driving cycles as the Worldwide Harmonized Light-Duty Vehicle Test Cycle (WLTC) within a detailed SiL environment. By considering all relevant influencing factors in a dynamic simulation, the method provides a more accurate assessment of fuel consumption and emission differences between two versions of the transmission software. The significant contribution of this work lies in the high-fidelity integration of a real virtual Transmission Control Unit (vTCU) software within a comprehensive, validated forward-looking SiL environment. This approach enables precise delta comparisons that capture transient dynamic interactions, facilitating early, reliable software testing and validation.
Kengne Dzegou, Thierry JuniorSchober, FlorianRebesberger, RonHenze, Roman
The calibration of automotive electronic control units is a critical and resource-intensive task in modern powertrain development. Optimizing parameters such as transmission shift schedules for minimum fuel consumption traditionally requires extensive prototype testing by expert calibrators. This process is costly, time-consuming, and subject to variability in environmental conditions and human judgment. In this paper, an artificial calibrator is introduced – a software agent that autonomously tunes transmission shift maps using reinforcement learning (RL) in a Software-in-the-Loop (SiL) simulation environment. The RL-based calibrator explores shift schedule parameters and learns from fuel consumption feedback, thereby achieving objective and reproducible optimizations within the controlled SiL environment. Applied to a 7-speed dual-clutch transmission (DCT) model of a Mild Hybrid Electric Vehicle (MHEV), the approach yielded significant fuel efficiency improvements. In a case study on a 4.7 km Worldwide harmonized Light-Duty vehicles Test Cycle (WLTC) driving segment, the RL-optimized shift strategy reduced fuel consumption from a baseline of 0.46 L to 0.37 L. Furthermore, when starting from an already optimized shift map representative of a series production vehicle’s calibration, the artificial calibrator further enhanced fuel efficiency, achieving approximately a 0.6 % reduction in fuel consumption for the 4.7 km segment and nearly a 5 % reduction for the full WLTC. The artificial calibrator thus demonstrates a promising methodology to frontload calibration tasks in simulation, thereby offering the potential to reduce reliance on resource-intensive physical testing and to significantly accelerate the development of fuel-efficient powertrain control software.. The direct compatibility of parameter files with real vehicle Electronic Control Unit (ECUs) and the validated SiL behavior suggest high transferability of learned strategies, offering the potential for minimal fine-tuning on physical vehicles post-simulation.
Kengne Dzegou, Thierry JuniorSchober, FlorianRebesberger, RonHenze, Roman
With the advancement of the automotive industry, more sophisticated systems are being incorporated into vehicles to enhance performance, safety, and other essential features. As a result, the software to control these complex systems continue to grow and complex. There is a need to test all these software thoroughly in a systematic way to ensure the correctness of implementation and functionality. Till now, test cases for the automotive software unit feature are being created manually and turns out to be inefficient and time consuming. There is a pressing need, therefore, to generate the testcases automatically based on the test requirements. To address this challenge, this research illustrates the use of large language models (LLMs) in automated test case generation for MIL/SIL/PIL platform. Recently, LLMs have been effectively applied to address challenges in natural language understanding, text generation, code generation, and more. Following this concept, our approach is to first create a structured document from the test requirements, utilizing the Llama LLM model, which is guided by prompt engineering. This structured document then used to create the testcases automatically using scripts. The performance of our approach is demonstrated with several examples including both static and dynamic scenarios. Additionally, we show the results on the publicly available test specification document from the Ministry of Road Transport and Highways, India. We present a comparative analysis of automatically versus manually generated test cases, showing that automation reduces effort by approximately 67%, completing tasks in one-third the time required for manual creation. Our approach of using LLM is seen to be more efficient generating test cases in a faster way for both static and dynamic scenarios. Further, our approach produces consistent and reproducible results managing different paraphrasing of the same test requirements.
Bagari, MayankPrajapati, SauravTuladhar, YunishKoti, Akshay
Nowadays, Software-in-the-Loop (SIL) represents a crucial methodology in the development and validation of control systems, particularly in sectors such as automotive, marine, and aerospace. It involves creating a virtual representation of a real environment with varying levels of accuracy. Using SIL techniques, engineers can develop and test software in the early stages of the development cycle, reducing overall time-to-market and costs. Typically, to simulate complex control systems, a primary tool is used to manage and integrate an entire application-specific environment composed of application software, plants, sensors and actuators, and communication protocols. Although several commercial solutions are currently available on the market to support SIL activities, Dumarey Softronix wanted to explore the possibility of developing an in-house software tool to leverage the benefits of SIL. This paper provides a high-level overview of the main steps involved in developing a complete SIL framework in contexts compliant with Autosar and based on Functional Mock-up Interface (FMI) standards. In the first part of the paper, the main choices behind the development of a SIL tool will be presented, such as the selection of the GUI framework selection, the simulation engine and the co-simulation framework. In the second part, the authors will discuss the key features a SIL solution should offer, including C-code building, C-code debugging, calibration loader, XCP/ASAP2, and a bus analyzer. In the final part of the paper, the tool is assessed by selecting two projects and comparing the results of a HIL bench with SIL results. The comparison highlights the consistency and accuracy of the SIL tool, demonstrating its ability to replicate real-world scenarios in a virtual simulation environment.
Mancuso, ClaudioTesconi, CristianAutieri, Fabio
Bloomy: COTS HIL/SIL/COPPER Bird Testing Equipment
Blume, Peter
In the pursuit of customizability and evolvability of vehicle functions, manufacturers shift towards software-defined vehicles to enable flexible customization and over-the-air updates. This results in multiple variants and versions of a vehicle model. While shifting to software-defined vehicles (SDVs) adds value and flexibility for customers, manufacturers struggle with homologating new and updated functionality because existing testing processes do not scale for high-frequency release cycles that limit available testing resources. Overcoming this challenge by using a coherent test process designed for testing continuously evolving variant-rich systems will be one of the key enablers. This paper presents an innovative end-to-end pipeline for efficient and comprehensive testing of variant-rich vehicle functionality tailored to an application in continuous development. Our transferable test pipeline employs sample-based variant selection, a software-in-the-loop environment for executing selected variants, and scenario-based testing using mutation-based scenario fuzzing. A central test controller is responsible for managing the process. We evaluate our test pipeline with regard to its fault detection capability, using the YOLO object detection algorithm as a variant-rich test object in the CARLA simulator. Our results show that the testing process outperforms random variant sampling and scenario mutation in detecting faults.
Hettich, LennardPett, TobiasNägele, Ann-ThereseSchindewolf, MarcEriş, HalitWagner, StefanSax, EricSchaefer, InaWeyrich, Michael
This study investigates the fault tolerance of a large-scale coaxial quadrotor Electric Vertical Takeoff and Landing (eVTOL) under motor failure through high-fidelity software-in-the-loop (SIL) simulations using PX4-Gazebo environment. The objective is to evaluate the vehicle's ability to maintain flight stability and complete critical missions under various propulsion failure scenarios, without the control system being explicitly aware of which motors have failed. Four motor failure cases-single, two adjacent, two diagonally opposite, and three distributed motor failures-were introduced during takeoff, hover, cruise, and hover under crosswind missions. Results show that the eVTOL maintained controllability and mission completion under all scenarios, with increasing levels of performance degradation under more severe failures. Notably, considerable yaw instabilities of about 10 degrees occurred under two diagonally opposite motor failures. The highest thrust demands after motor failures were observed during cruise mission, with some motors demanding about 80% to 90% of their maximum throttle. Hover under crosswind revealed compounded challenges in attitude control during descent under severe failure cases compared to calm weather. These findings underscore the robustness of the integrated control system and vehicle configuration in managing motor failure scenarios.
Asadi Khanouki, MostafaSadat-Nejad, YounesPourmostaghimi, Nima
Less costs and higher efficiency may be constant technological pursuit. Despite the great success, data-driven AI development still requires multiple stages such as data collection, cleaning, annotation, training, and deployment to work together. We expect an end-to-end style development process that can integrate these processes, achieving an automatic data production and algorithm development process that can work with just clicks of the mouse. For this purpose, we explore an end-to-end style parking algorithm development pipeline based on procedural parking scenario synthetic data generation. Our approach allows for the automated generation of parking scenarios according to input parameters, such as scene construction, static and dynamic obstacles arrangement, material textures modification, and background changes. It then combines with the ego-vehicle trajectories into the scenarios to render high-quality images and corresponding label data based on Blender software. Utilizing Blender as render engine is one key part of our pipeline as its film-level high fidelity results guarantee the data transfer performance in real world, which makes Blender superior to solutions with game engines for parking scenario generation. We conduct experiments based on our own vehicle, automatically generated 100 parking scenes, combine expert controllers on the Software In-the-Loop platform and offline optimization algorithms to generate 1000 parking trajectories, and render more than 200000 frames end-to-end parking data. Finally, the generated data is fed into the end-to-end parking algorithm and applied in a real vehicle to park it into specific parking-slots. The experimental results demonstrate that our goal has been basically achieved, with only mouse clicks to develop a parking function.
Li, JianWang, HanchaoZhang, SongMeng, ChaoRui, Zhang
Energy management strategy is essential for HEV’s to achieve an optimum of energy consumption. With predictive energy management, taking future vehicle speed predicted from ADAS map information, in-vehicle navigation traffic flow status information, and current speed into account, one could anticipate a considerable improvement in energy-saving. The major validating approach widely adopted for energy management algorithms nowadays is real-world vehicle testing, of which the economic and time costs are relatively high. Moreover, with advanced algorithms featuring AI coming into light, putting forward higher requirement in the richness of test cases, the drawback in coverage of vehicle testing is revealed. This paper proposed a MIL/SIL testing approach for predictive energy management algorithms, providing a partial replacement to, and overcome the limitations of, vehicle testing. In the testing setup, random traffic generated by MATLAB® based on real-time traffic condition will be taken over by SUMO [15]. In simulation, a map sensor fetches required signals at pre-sampled feature points on the planned trajectory. The collected data include road slope, speed limits, traffic flow densities, average speeds, etc., within which the static data will be provided by a high-fidelity map hosted in RoadRunner, runtime information is computed from the status of target vehicles on-the-fly. This testing approach can, in algorithm validation, not only save cost, but also offer the possibility of scenario variation, therefore enriched test case base, and set the foundation for further analysis on impact of performance of different factors.
Yan, YueMa, XiudanWei, XinliXiong, JieDeng, YunfeiBradfield, Donald Edward
Bloomy: HIL/SIL Verification and Validation Testing
Blume, Peter
In Automotive world, vehicle development includes design and testing of hardware and software. Hardware includes components required for actuation and sensing, along with the controller hardware. Software includes control logic embedded in controller for functioning of these components. Generally, software inside controller could be validated in various ways e.g., Software in Loop (SiL), Hardware in Loop (HiL), Vehicle testing. During initial phase of control software development cycle, plant models with adequate accuracy replicating hardware components are utilized for digital software validation. Many a times, hardware components might be available before control software matures. Hence, to validate plant models for their accuracy & quality alternate option of actual controller is needed during initial phase. Intelligent controller mimicking original controller can be an alternate option for plant model improvement and component level performance analysis. This paper proposes a Reinforcement Learning (RL) based intelligent controller for controlling Fan actuator of thermal management system of an electrical vehicle. In simulation world, this becomes advantageous from resource availability point of view, especially during initial phase of control software development. The same methodology can be extended to other components of system as well.
Chhagar, Rohnit SinghNavse, SiddharthKumar, Lavanya
Advanced Driver Assistance Systems (ADAS) is a growing technology in automotive industry, intended to provide safety and comfort to the passengers with the help of variety of sensors like radar, camera, Light Detection and Ranging (LIDAR) etc. The camera sensors in ADAS used extensively for the purpose of object detection and classification which are used in functions like Traffic sign recognitions, Lane detections, Object detections and many more. The development and testing of camera-based sensors involves the greater technologies in automotive industry, especially the validation of camera hardware and software. The testing can be done by various processes and methods like real environment test, model-based testing, Hardware, and Software in loop testing. A fully matured ADAS camera system in the market comes after passing all these verification processes, yet there are lot of new failures popping up in the field with this ADAS system. Since ADAS is an evolving technology, many new field issues keep coming due to huge diversity of features in the real-world infrastructure. So, to bring up a more reliable ADAS system, validating every newly reported issue and including the fix in the software is the only way to bring the safe and reliable system in the Automobiles. If there is any issue reported in certain places, the current practice is to reproduce the issue in the same place, analyzing the issue & finding the solution with fix in the software and finally validating that issue in the same location. This process of reproducing the issue & validating the fix in the same location can become more complex and expensive since that hotspot locations can be found anywhere in the world. This paper proposes a technique for the camera-based testing includes maps-based testing by using real maps scenarios played in front of the camera in controlled simulation environment and evaluate the results to confirm the software maturity. Real world scenarios can be created using open-source maps data and using it after fine tuning in the Hardware in Loop (HiL) system can reduce the complexity and cost of development & validation of ADAS camera Sensors.
R, ManjunathSaddaladinne, JagadeeshPachaiyappan, Sathish
Electrified drives will change significantly in the wake of the further introduction of automated driving functions. Precise drive dimensioning, taking automated driving into account, opens up further potential in terms of drive operation and efficiency as well as optimal component design. Central element for unlocking the dimensioning potentials is the knowledge about the driving functions and their application. In this paper the implications of automated driving on the drive and component design are discussed. A process and a virtual toolchain for electric drive development from concept optimization to detailed dimensioning validation is presented. The process is subdivided into a concept optimization part for finding the optimal drive topology and layout and a detailed prototype environment, where more detailed component models can be assessed in customer operation to enable representative component dimensioning. Furthermore, the detailed simulation allows the drive investigation in representative customer operation as well as automated driving functions in terms of a software in the loop simulation. The process is used for the optimal dimensioning of a battery electric vehicle of the D-segment. The work focusses on a highway pilot function, developed at the Institute of Automotive Engineering of the Technische Universität Braunschweig. The optimal drive configuration can later be transferred to the prototype dimensioning. The simulation of automated driving function operation is based on a vehicle following scenario which employs statistical human behavior in the target vehicle and a sliding mode ACC in the ego vehicle. This methodology is particularly suitable for determining load spectra, which in turn can be used as test specifications for the strength simulation or endurance testing of the electric drive. Furthermore, simulation results can be used for the definition of representative cycles applicable for the concept optimization. The results of both processes will be compared and discussed in detail with an emphasis on efficiency, performance and load spectra.
Sturm, Axel WolfgangBrandes, GerritSander, MarcelHenze, RomanKüçükay, Ferit
The modern automotive industry is facing challenges of ever-increasing complexity in the electrified powertrain era. On-board diagnostic (OBD) systems must be thoroughly calibrated and validated through many iterations to function effectively and meet the regulation standards. Their development and design process are more complex when prototype hardware is not available and therefore virtual testing is a prominent solution, including Model-in-the-loop (MIL), Software-in-the-loop (SIL) and Hardware-in-the-loop (HIL) simulations. Virtual prototype testing relying on real-time simulation models is necessary to design and test new era’s OBD systems quickly and in scale. The new fuel cell powertrain involves new and previously unexplored fail modes. To make the system robust, simulations are required to be carried out to identify different fails. Thus, it is imminent to build simulation models which can reliably reproduce failures of components like the compressor, recirculation pump, humidifier, or cooling systems. This paper shows the development of high-fidelity fuel cell model which is used as digital twin to reproduce relevant failure modes. As the OBD regulations become more stringent and advanced, it is difficult to keep pace with it and perform comprehensive testing in real world environment. In such scenarios, MIL, SIL and HIL testing becomes more prevalent. MIL and SIL testing provide a quick way for controls engineers to develop new strategies at system level to adhere to new OBD regulations. On the other hand, simulating high fidelity physics based Real Time plant model on HIL systems, allows the engineers to perform fault insertions tests on the software and leave the lab environment with a certain degree of guarantee that the software would fare well in real world conditions. The model used can reproduce failure modes consistently while staying in real time which in turn can be detected by controls and can take action promptly. The viability of this approach is demonstrated by showing MIL and HIL test results.
Pandit, Harshad RajendraDimitrakopoulos, PantelisShenoy, ManishAltenhofen, Christian
Validation plays a crucial role in any Electronic Development process. This is true in the development of any automotive Electronic Control Unit (ECU) that utilizes the Automotive V process. From Research and Development (R&D) to End of Line (EOL), every automotive module goes through a plethora of Hardware (HW) and Software (SW) testing. This testing is tedious, time consuming, and inefficient. The purpose of this paper is to show a way to streamline validation in any part of the automotive V process using Python as a driving force to automate and control Hardware-in-the-loop (HIL) / Model-in-the-loop (MIL) / Software-in-the-loop (SIL) validation. The paper will propose and outline a framework to control test equipment, such as power supplies and oscilloscopes, load boxes, and external HW. The framework includes the ability to control CAN communication signals and messages. A visual Graphical User Interface (GUI) has also been created to provide simplified operation to the user without knowledge of the backend. This provides a user-friendly display to synchronize multiple SW and HW toolsets together. The proposed methodology creates an efficient automation suite that can be tailored to any type of validation. The GUI provides portability, diversity, and simplicity, which allows for a streamline and efficient process for validation. The paper will refer to user cases to demonstrate the process and setup to perform functionality followed by the conclusion.
Rosiewicz, BrandonLink, Bravin
The rise of Software-Defined Vehicles (SDV) has rapidly advanced the development of Advanced Driver Assistance Systems (ADAS), Autonomous Vehicle (AV), and Battery Electric Vehicle (BEV) technology. While AVs need power to compute data from perception to controls, BEVs need the efficiency to optimize their electric driving range and stand out compared to traditional Internal Combustion Engine (ICE) vehicles. AVs possess certain shortcomings in the current world, but SAE Level 2+ (L2+) Automated Vehicles are the focus of all major Original Equipment Manufacturers (OEMs). The most common form of an SDV today is the amalgamation of AV and BEV technology on the same platform which is prominently available in most OEM’s lineups. As the compute and sensing architectures for L2+ automated vehicles lean towards a computationally expensive centralized design, it may hamper the most important purchasing factor of a BEV, the electric driving range. This research asserts that the development of dynamic sensing and context-aware algorithms will allow a BEV to retain energy efficiency and the ADAS to maintain performance. Moreover, a decentralized computing architecture design will allow the system to utilize System-on-Module (SoM) boards that can process Artificial Intelligence (AI) algorithms at the edge. This will enable refined hardware acceleration using Edge-AI. The research will propose the use of a novel Software-in-the-Loop (SiL) simulation environment for a 2023 Cadillac LYRIQ provided by the EcoCAR EV Challenge competition. Future work will involve an in-depth evaluation and discussion of the simulation data. We will conclude that optimizing sensing and computation in an SDV platform will allow Automated and Electric Vehicles to prosper concurrently without impeding their technological progress.
Kothari, AadiTalty, TimothyHuxtable, ScottZeng, Haibo
In autonomous technology, uncrewed aircraft systems have already become the preferred platform for the research and development of flight control systems. Although they are subjected to following and satisfying complicated scenarios of control stations, this high dependency on a specific control framework limits them in their application process and reduces the flight self-organizing network. In this article, we present a developed multilayer control system protocol with the additional supportive manned aircraft layer (Tender). The novelty of the introduced model is that uncrewed aircraft systems are monitored and navigated by the tender, and then based on the suggested scheme, data flows are controlled and transferred across the network by the developed cloud–robotics approach in the ground station layer. Therefore, it has been tried to design a semi-autonomous control network to gather data that combines human observation and the automotive nature of uncrewed aircraft systems. To ensure the accuracy and correctness of the model, we simulate our approach in the software-in-the-loop using its web-based interface with new configurations in the hardware and software architecture of the network. Results will be examined by the in-order per message delay, which has recorded a considerably low latency in both the uplink and downlink data transmission processes. This optimization is achieved along with maintaining the quality of data.
Millar, Richard C.Laliberté, JeremyMahmoodi, ArminHashemi, LeilaMeyer, Robert Walter
In today’s scenario, the software validation phase in the automotive software development cycle uses different testing environments/platforms (Verification uses Model in Loop (MiL), function validation uses Software in Loop (SiL), HW and system level tests uses Hardware in Loop (HiL), and system testing uses vehicle). Each of these platforms are highly expensive to deploy and maintain and poses significant constraints in directly comparing the test results across platforms due to the difference in test reporting structures. It is also noticed that there is high redundancy or overlapping of test cases and timely availability of these platforms for timely SW release testing. The solution helps in integrating MiL, SiL, and HiL test platforms into one single "Integrated Testing Platform," which eventually saves testing time, effort, and cost, along with early bug detection during the software development phase. With the reuse of relevant compatible test cases in each of the test phases, test-case redundancy is eliminated, thus reducing total man-hours spent in creating similar tests in different environments at different phases for the same function. Additional benefits include improved test-case coverage and fault reaction testing on digital platforms. Each of the platforms are compatible with a single reporting tool (e.g., CANape logs), thus streamlining the test report format. In this manner, HiL and vehicle test platforms can be dedicated to hardware/system-specific tests, and application layer logic can be tested via this platform during the early stage of software development. The closed-loop digital validation platform includes plant model, SiL, and HiL connected with each other via Silver [3] and PROVEtech [2] tool wherein PROVEtech [2] acts as a master for controlling and Silver [3] serves as the backbone for all the test simulations. The results from this platform have a standard format and thus can be compared and analyzed via a standard analysis tool. This helps in easy reporting structure across different platforms.
Jain, NayankN T, SavithaG, SudipBhogenahally Lakshminarayana, AnuroopaManish, Kumar
The evolution of automotive Electronic Control Unit (ECU) technology brings the additional safety, comfort, and control to the vehicle. With an exponential increase in the complexity involved in modern-day ECU, it is very important to verify and validate robustness, functionality, and reliability of ECU software [1]. As of now, Hardware in loop [HIL] and Vehicle in Loop validations are well known software functional validation methods. However, these methods require physical setup, which can incur more cost and time during the development phase. In recent years, ECU virtualization gained attention for development and validation of automotive ECUs [2]. The goal is to minimize the effort on software testing. This paper focuses on virtualization of Electric Vehicle (EV) powertrain system using SIL approach. The objective is to provide an adaptable EV-virtualization environment for virtual-ECU (vECU) verification and validation. This paper focuses on standardization of SIL simulation setup by using modularized plant models. The term virtualization refers to a methodology of simulating automotive ECUs in a virtual environment (IT equipment) using SIL approach. In this virtual validation approach, ECU software stack is compiled and converted into an executable which can run on general purpose IT equipment with Windows or Linux environment. This executable is called a vECU. As there are no hardware dependencies, virtual ECUs provide the advantage of testing various scenarios in a simulated environment [3]. The demonstrated EV virtualized SIL solution could ultimately result in faster and more cost-effective software development cycle.
Sajnani, AbhishekVernekar, KiranGosavi, RupeshNaik, Venkatesh
A previously developed piston damage and exhaust gas temperature models are coupled to manage the combustion process and thereby increasing the overall energy conversion efficiency. The proposed model-based control algorithm is developed and validated in a software-in-the-loop simulation environment, and then the controller is deployed in a rapid control prototyping device and tested online at the test bench. In the first part of the article, the exhaust gas temperature model is reversed and converted into a control function, which is then implemented in a piston damage-based spark advance controller. In this way, more aggressive calibrations are actuated to target a certain piston damage speed and exhaust gas temperature at the turbine inlet. A more anticipated spark advance results in a lower exhaust gas temperature, and such decrease is converted into lowering the fuel enrichment with respect to the production calibrations. Moreover, the pollutant emissions associated with production calibrations and the implementation of the developed controller are compared through a GT-Power combustion model. Finally, the complete controller is validated for both the transient and steady-state conditions, reproducing a real vehicle maneuver at the engine test bench. The results demonstrate that the combination of an accurate estimation of the damage induced by knock and the value of the exhaust gas temperature allows to reduce the brake specific fuel consumption by up to 20%. Moreover, the stoichiometric area of the engine operating field is extended by 20%, and the GT-Power simulations show a maximum CO reduction of about 50%.
Brusa, AlessandroMecagni, JacopoShethia, Fenil PanalalCorti, Enrico
The Virtual Autonomous Navigation Environment (VANE) is a set of tools that have been developed over a decade to assist autonomy developers in building autonomous systems. VANE has high-fidelity, physics-based sensors and vehicle models that interact with virtual environments built by utilizing decades of experience in characterizing environmental conditions. These models and environments are used in software-in-the-loop simulations to assist in the development and evaluation of autonomous vehicles in a cost-effective and time-sensitive manner. The software-in-the-loop simulations have been verified with data from concurrent physical testing and are used by autonomy developers to improve the safety, scalability, and cost effectiveness of testing autonomous vehicles.
Holden, GarrettAspin, ZacharyMonroe, John G.McInnis, DavidDavenport, CollinPrice, PhillipHansen, Brad
Automated driving, electrification, cloud computing and the push toward software-defined vehicles are forcing automotive and commercial-vehicle developers to revamp design strategies. Tools suppliers are moving to help engineers develop and verify solutions that address the complete vehicle environment, a task that requires a growing number of design tools. During the recent dSPACE World Conference in Munich, Germany, several vehicle manufacturers described their strategies for coping with these trends. dSPACE, which supplies hardware/software-in-the-loop (HIL/SIL) tools, announced plans to see if tool makers can find a way to help developers by making it easier to integrate data created using different development software.
Costlow, Terry
The front camera module is a fundamental component of a modern vehicle’s active safety architecture. The module supports many active safety features. Perception of the road environment, requests for driver notification or alert, and requests for vehicle actuation are among the camera software’s key functions. This paper presents a novel method of testing these functions virtually. First, the front camera module software is compiled and packaged in a Docker container capable of running on a standard Linux computer as a software in the loop (SiL). This container is then integrated with the active safety simulation tool that represents the vehicle plant model and allows modeling of test scenarios. Then the following simulation components form a closed loop: First, the active safety simulation tool generates a video data stream (VDS). Using an internet protocol, the tool sends the VDS to the camera SiL and other vehicle channels. The camera SiL performs its functions (e.g., object detection, active safety functions) and provides output (e.g., braking request) to the active safety simulation tool. The vehicle plant model updates the vehicle state (e.g., applies braking). The vehicle’s position is updated in the context of the test scenario, and a new VDS frame is generated. Preliminary results are presented and discussed along with potential uses of this simulation setup in vehicle active safety development.
Elbaz Elsaiid, MoatazKalliman, SamuelKral, Jiri
The accelerated processes in vehicle development require new technologies for function development and validation. With this motivation, Function-in-the-Loop (FiL) simulation was developed as a link between Software-in-the-Loop (SiL) and Hardware-in-the-Loop (HiL) simulation. The combination of real Electronic Control Unit (ECU) hardware and software in conjunction with virtual components is very well suited for function development and testing. This approach opens up new possibilities for mechatronic systems that would otherwise require special test benches. For this reason, an Electric Power Steering (EPS) was transferred to a virtual environment using FiL simulation. This enables a wide range of applications, from EPS testing to the development of connected driving functions on an integrated platform. Right from the early development phases, the technology can be used purposefully with short integration cycles. Throughout the entire development process, function development and validation can be effectively controlled and quality increased.
Achilles, FrederikSteib, FrederikNippold, ChristophHenze, Roman
An autonomous vehicle is able to perceive and interpret exactly its surroundings and its interior (“Sensing”). then, it processes the information received and plan its driving strategy (“processing”). And finally, it uses its powertrain, steering and braking power to move its wheels in such a way that the planned driving strategy is put into practice (“Acting”). Testing an autonomous vehicle’s reaction to the erratic traffic scenarios using prototypes would be impractical. Physically testing these scenarios can also be risky to human life and equipment. Additionally, the repetition involved in the comprehensive testing of all these scenarios could lead to human errors. Various Self Driving car manufacturers have reported injuries and causalities while doing Functional testing [1]. Testing autonomous vehicles with simulations can model faulty sensors to determine whether the autonomous vehicle is functionally safe and also provides the comprehensive test coverage by saving time to market and also the cost. Hence a Virtual Simulation Platforms are needed. ROS 2 (Robot Operating System 2) is an open-source software development kit for robotics applications which is specifically built to allow for plug and play packages independent of the software stack. The purpose of ROS2 is to offer a standard software platform to developers across industries that will carry them from research and prototyping through to deployment and production. This Open-source ROS has been growing exponentially [2]. In ADAS simulation Environments like SIL, HIL, PIL that are based on ROS2 using rmw_fastrtps as Abstract DDS layer, it just isn’t fast enough and less reliable also. Hence there is need for a reliable and low latency communication protocol at the ROS2 middleware. This paper presents a comparative study between ROS2 Abstract layers like rmw_fastrtps and ecal_rmw with respect to reliability and latency aspects.
Venkannacharya, Gururaj
Autonomous vehicles (AVs) are self-contained vehicles equipped with control systems to execute various tasks. The Lane Departure Warning (LDW) system is widely employed to prevent the most common cause of vehicle collisions. An autonomous lane-departure system will aid and reduce such collisions. When the vehicle is at risk of drifting or departing its lane, the LDW system monitors its relative position to the lane edge and sends an alarming warning signal to the driver. This work uses an ML-based technique to detect lane markers in an Indian context using a high-resolution camera mounted on the car. Considering that, the LDW system requires three primary operations. The camera geometry information is used to divide the acquired image into two parts: a road part and a non-road component. Then, to circumvent the obstacles caused by the perspective effect, inverse perspective mapping is applied. Then, using a sliding window technique, lane markers are filtered, and Canny edge detection is performed. In addition, the Hough transform method is employed to detect lane markers. If the system detects the vehicle is too close to the right, or left lane markings, a warning light, vibration, or sound will be activated, according to the euro NCAP standard for a lateral offset of 0.76 m. TiHAN IITH test track evaluates ADAS functionality for the Indian market as per euro NCAP standard. In addition, a TiHAN IITH test track was built and validated using real-time testing with ADAS capabilities using the SIL-based software environment. Many sensors in the vehicle dynamics of the mid-size M1 automobile have been adapted to communicate with the SIL environment. For experimental validation, SIL provides iterations of repeated actions for test scenarios that pass or fail based on the NCAP test score.
Verma, Amar KumarR, VaibhavPerabhattula, VenkateshRajalakshmi, P
Simulation of real time situations is a time tested software validation methodology in the automotive industry and array of simulation technologies have been in use for decades and is widely accepted and been part & parcel of software development cycle. While software that is being developed needs detailed plan, architecture and detailed design, it also matters during its development that, it is built in the right way from the very beginning and is fine tuned constantly. Especially for Software-In-Loop simulation (SIL), plenty of practices/tools/techniques/data are being used for simulation of system/software behavior. When it comes to choosing the right simulation technique and tools to be adopted, often there are discussions revolve around cost, feasibility, effectiveness, man-power, scalability, reusability etc. As automotive software validation is data driven, we deal with myriad of ground truth data for simulations, ranging from vehicle dynamics to vehicle models to environment factors (road, test track, weather). While infusing the ground truth data (physical data) in a simulation environment is still possible, availability of the same is often sparse, owing to test track constraints, availability of target objects, high risk maneuvers (involving higher vehicle speeds, narrow road curvatures, weather factors, accident reconstruction etc.) To tide over these constraints, in recent times, organization are inclined towards adopting virtual software simulation techniques, as it offers precision, scalability, cost and time effectiveness, and helps developers with early feedback to fine tune the software. CarMaker (Third Party Simulation tool) covers all the benefits stated above when it is integrated with the software (to be tested) in a closed loop. Vehicle data, Sensor data and real time scenarios (use cases) can be readily modelled for customer requirements for a product/SW and simulated and tested readily to provide to the developers on system/SW performance, ranging from ego dynamics, trajectory, camera/radar calibration, component/subsystem performance, accident severity, accident implications dealing with costs & damages, impact velocity, collision mitigation etc. Closed loop simulations are widely in use during development and testing of safety/cruising applications like ACC (Adaptive Cruise Control), EBA (Emergency Braking Assist), LCF (Lateral Control Functions) and also assists in functional areas of NCAP (New Car Assessment Programme) [1] and GIDAS (German In-Depth Accident Study) topics.
Nagarajan, KalaiyarasanRanga, AnkurKalkura M, KiranAnegundi, RanishreeAriharan, Anantharaju
In this article, a formation flying technique designed for a multiple unmanned aerial vehicles (multi-UAV) system to provide low-cost and efficient solution for civilian and military applications is presented. First, a modular leader-follower formation algorithm was developed to accomplish the formation flying with off-the-shelf low-cost components and sensors. Second, a proportional-integral-derivative (PID) controller was utilized for velocity control of the UAVs to maintain the tight formation. Third, a particle swarm optimization-optimized reciprocal velocity obstacles (PSO-RVO) algorithm was utilized for obstacles avoidance and collision avoidance between the UAVs while navigating, with the aid of sonar ranging sensors onboard. The formation flying algorithm developed was tested through both simulation and experiment using two quadcopters with global positioning system (GPS) signals. For the simulation, the algorithm developed was tested on a virtual quadcopter using an open-source software-in-the-loop (SITL) simulator. With the aid of the experimental test, the effectiveness of the proposed formation flying algorithm is evaluated. With a separation distance of 5 m between the UAVs, the proposed system is able to achieve an average separation error of 0.3872 m and percentage of root mean square error (RMSE) of 9.7%. Therefore, it is shown that the proposed formation flying system is very effective.
Cheok, Jun HongAparow, Vimal RauNg Zhi Neng, JunoCheah, Jian LeeLeong, Dickson
Due to the ever increasingly stringent emission regulations for passenger vehicles, the efficiency and performance increase of Spark Ignition (SI) engines have been under the focus of the engine manufacturers. The quest for efficiency and performance increase has led to the development of increasingly complex powertrains and control strategies. The development process requires novel methods that feature a smooth transition between the real and the virtual prototypes. Furthermore, to reduce the development time and cost, developing an engine simulator with a low computational effort and good accuracy, which predicts the engine behavior on the entire operating range, plays a crucial role. This work proposes an Artificial Intelligence-based engine simulator for a Spark Ignition engine. The simulator relies on Neural Networks for the calculation of the main combustion metrics. In the first part of this paper, the data acquired at the engine test cell are analyzed. A shallow neural network model is set up in Matlab for modelling the combustion phase and the knock intensity. The training of the network is performed with a Design of Experiment (DoE) approach, where different numbers of neurons are tested using training algorithms and activation functions. The output of the simulation is then compared to the experimental values. The results are evaluated using the Root Mean Square Error and R-square indexes, and the combination of the number of neurons, training algorithm, and activation function, which yield the lowest error and the highest R-square, are selected. In the second part, the ANN models are then coupled to analytical sub-models previously developed and individually validated, to build a complete engine simulator, and they are implemented in Simulink. The performance of this simulator is then evaluated by comparing simulated results and experimental data, both under steady-state and transient conditions.
Shethia, Fenil PanalalMecagni, JacopoBrusa, AlessandroCavina, Nicolo
Spurred by the constraints of the COVID-19 pandemic, virtual testing is becoming an increasingly essential method for verification and validation of autonomous ground vehicle simulation tools. The Mobility Systems Branch (MSB) of the US Army Corps of Engineers Engineering Research and Development Center (ERDC) Geotechnical and Structures Laboratory (GSL) has developed a new approach in physics-based virtual testing of autonomous ground vehicle systems through the incorporation of both qualitative and quantitative data in congruency with ERDC’s Software-in-the-Loop laboratory. Virtual testing of autonomous vehicles combines simulation tools consisting of vehicle and sensor models represented in a virtual scene with both performer observations and modeling and simulation observations. The first iteration of a Virtual Engineering Evaluation Test (V-EET) for robotic and autonomous ground vehicle systems took place in 2021 at the ERDC in Vicksburg, Mississippi. Virtual testing took place over the course of several months with remote researchers participating from across the country. Researchers used a combination of quantitative and qualitative methods to identify discrepancies between traditional field Engineering Evaluation Test data collection and V-EET data collection. Also identified were issues within testing protocols and difficulties associated with overwhelming and complex data sets. Building on findings, researchers developed a new virtual testing framework that addressed these issues and included more versatility. This new framework included streamlined and efficient data collection and analysis, standardization of observation collection techniques, and objectification of qualitative data to be used across relevant products with visual or human components. This will then provide the most efficient and robust products possible and improve situational awareness for autonomous vehicle assessment in complex on- and off-road environments.
Lyons, JessicaJackson, RebekahRichards, JamesGates, BurhmanFairley, JoshuaPrice, Stephanie
Software-in-the-Loop (SiL) test environments are the ideal virtual platforms for enabling continuous-development, -integration, -testing -delivery or -deployment commonly referred as Continuous-X (CX) of the complex functionalities in the current automotive industry. This trend especially is contributed by several factors such as the industry wide standardization of the model exchange formats, interfaces as well as architecture definitions. The approach of frontloading software testing with SiL test environments is predominantly advocated as well as already adopted by various Automotive OEMs, thereby the demand for innovating applicable methods is increasing. However, prominent usage of the existing monolithic architecture for interaction of various elements in the SiL environment, without regarding the separation between functional and non-functional test scope, is reducing the usability and thus limiting significantly the cost saving potential of CX with SiL. In this paper, we introduce a novel modular architecture of SiL environments especially suitable seamlessly for such CX use cases, considering various functional and non-functional test scopes. Particularly in section 1 of this paper, we elaborate the state-of-the-art standards in the industry and use cases of SiL from the OEM and supplier perspective. Then we demonstrate the benefits modular SiL environments compared to the monolithic counterparts, for different test scopes in section 2 and illustrate CX pipeline with these environments in section 3. Finally, we conclude with an elaborate list of key performance indicators (KPIs) and establishing future scope.
Raghupatruni, IndrasenKarjee, SambuddhaGupta, AnupamNaik, VenkateshHuber, Thomas
Model Release Process using Standardized Error Metrics for Validation of X-in-the-Loop Simulation Models2021-01-11489/21/2021
The current automotive market is dynamic, leading to complex functionalities being incorporated into the control software of various components like engine, gearbox, battery, E-motor etc. This results in utilization of virtual environments for software testing to reduce the development time. The virtual platforms under the category X-in-the-Loop (XiL) e.g. Software-in-the-Loop (SiL) and Hardware-in-the-Loop (HiL) use simulated models to achieve a desired test goal. These component models must be rigorously validated to ensure the quality of XiL-Testing. Thus, it is essential to define a model release process that maintains model quality irrespective of the modeling approach used and the user. One of the challenges is to choose an appropriate Error Metric (EM) that sets criteria for model release. This paper proposes a combination of Theil’s Inequality Coefficient (TIC) and Unscaled Mean Bounded Relative Absolute Error (UMBRAE) as the EM. TIC is used to characterize the steady and non-dynamic variables. In contrast, UMBRAE is used to characterize transient dynamic variables. The entire operating range of the component model is validated using this combination of EM. This paper presents the process to validate an engine plant model against test-bench measurements. The quality level of each output signal is categorized as “excellent” if EM is less than 1, “good” if EM is less than 5, or “acceptable” if the EM is less than 10. The mean of EM of all signals represents the overall model quality and maximum of EM represents the worst-case scenario. The inaccurate signals and corresponding functionalities can be promptly identified through the process. These models can be iteratively improved to ensure that the quality demands are met for all the signals.
Narasimha, VedanthaNayak, ManishBick, Alexander
This paper presents the evolution of a series of connected, automated vehicle technologies from simulation to in-vehicle validation for the purposes of minimizing the fuel usage of a class-8 heavy duty truck. The results reveal that an online, hierarchical model-predictive control scheme, implemented via the use of extended horizon driver advisories for velocity and gear, achieves fuel savings comparable to predictions from software-in-the-loop (SiL) simulations and engine-in-the-loop (EiL) studies that operated with a greater degree of powertrain and chassis automation. The work of this paper builds on prior work that presented in detail this predictive control scheme that successively optimizes vehicle routing, arrival and departure at signalized intersections, speed trajectory planning, platooning, predictive gear shifting, and engine demand torque shaping. This paper begins by outlining the controller development progression from a previously published engine-in-the-loop study to the in-vehicle driver-in-the-loop testing highlighted in this work. The purpose of this field testing is to quantify the level of agreement between field data and prior results, particularly noting the influence of noise, disturbances and unmodeled dynamics. Also detailed are the steps to effectively implement the predictive horizons in experimentation with human rather than automated driver inputs by replacing the conventional speedometer with an augmented driver display. The driver display seeks to maintain driver awareness and route preview simultaneously. Additional features of the instrumented truck testing protocol include detailed route maps on speed limits, intersections, traffic and road grade; highly accurate fuel flowrate measurements calibrated via gravimetric methods; and closed-loop control between the predictive controller, driver, and vehicle controller area network (CAN). These combined capabilities allow for model validation, investigation of the effects of human-in-the-loop on controller performance, and identification of unmodeled dynamics and disturbances. The results show that the driver advisory implementation matches fuel consumption predictions from software-in-the-loop simulations and engine-in-the-loop studies to within 2-9%. Following the study of in-vehicle performance and variability, these results also validate, using real road tests, the up to 21% fuel savings achieved by using the hierarchical control scheme with a combination of multiple chassis and powertrain optimization technologies.
Pelletier, EvanBai, WushuangAlvarez Tiburcio, MiguelBorek, JohnBoyle, StephenEarnhardt, ChristianGao, LimingGeyer, StephenGraham, ChristopherGroelke, BenMagee, MarkPalmeter, KyleRodriguez, ManuelXu, ChuFathy, HosamNaghnaeian, MohammadStockar, StephanieVermillion, ChristopherBrennan, Sean
Over the years, the complexity of autonomous vehicle development (and concurrently the verification and validation) has grown tremendously in terms of component-, subsystem- and system-level interactions between autonomy and the human users. Simulation-based testing holds significant promise in helping to identify both problematic interactions between component-, subsystem-, and system-levels as well as overcoming delays typically introduced by the default full-scale on-road testing. Software in Loop (SiL) simulation is utilized as an intermediate step towards software deployment for autonomous vehicles (AV) to make them reliable. SiL efforts can help reduce the resources required for successful deployment by helping to validate the software for millions of road miles. A key enabler for accelerating SiL processes is the ability to use Simulation as a Service (SaaS) rather than just isolated instances of software. The primary benefits ensue from the in-parallel processing of multiple scenarios or tests using cloud or multiple cores especially to more systematically create “what-if analyses” thereby reducing both development time and cost. Here, we present the workflow of our utilization of SaaS methods (provisioned by Metamoto) and our explorations in this domain using exemplar ADAS scenarios. Additionally, we highlight our ability to perform parametric sweeps over variables such as environmental conditions, actors in the scene, etc. hence performing tests over a variety of scenarios including edge cases. The goal of our efforts is to examine viability and ease-of-use of SaaS (Metamoto in a co-simulation mode) to support Software-in-the-Loop co-development and functional reliability within MATLAB, ROS and Python frameworks.
Kagalwala, HuzefaSrivastava, SiddhantVenkatesan, Manikanda BalajiSrinivasan, SrivatsanKrovi, Venkat N
With the enhancements in vehicle electrification and autonomous vehicles, Traffic systems are also being improved at an accelerated rate to aid the development of improving fuel economy standards. For this to be possible, it is essential that traffic can be accurately modeled and predicted. The existing toolsets are proprietary and expensive and traffic modeling is not a trivial task due to its dependence on various factors such as place, time, and weather. To address these issues, an entirely open-source Software-In-Loop (SIL) fleet-focused traffic modeling toolset has been developed with the ability to take environmental factors with powertrain-in-the-loop into account leveraging Simulation of Urban Mobility (SUMO) and python. The proposed SIL toolset encompasses the creation of a microscopic traffic distribution which accounts for the usual traffic trends of a typical day. Parameters such as the number of vehicles entering the network and the speed of all the vehicles at a time of a day can be controlled by tunable weather conditions, which is obtained using weather APIs or datafiles. Given a network, origin and destination roads can be defined to create the routes based on shortest distance using default algorithms (like DUAROUTER) in SUMO on a pre-defined real network from Open Street Map or similar mapping standard. In this current study, the ego vehicle has been called into the traffic system after every hour, and information like battery charge, vehicle speed, and time of its trip are logged after every second at different traffic conditions. This developed framework can be used to test electric vehicles or fleets in a typical traffic scenario (which is determined by the weather conditions) and the effects of the traffic on the ego vehicle’s properties like the state of charge of the battery can be studied and modeled. This work can be extended to CAVs by using communication tools like Veins and OMNET++.
Padisala, Shanthan KumarYurkovich, Benjamin
Automated driving systems (ADS) are one of the key modern technologies that are changing the way we perceive mobility and transportation. In addition to providing significant access to mobility, they can also be useful in decreasing the number of road accidents. For these benefits to be realized, candidate ADS need to be proven as safe, robust, and reliable; both by design and in the performance of navigating their operational design domain (ODD). This paper proposes a multi-pronged approach to evaluate the safety performance of a hypothetical candidate system. Safety performance is assessed through using a set of test cases/scenarios that provide substantial coverage of those potentially encountered in an ODD. This systematic process is used to create a library of scenarios, specific to a defined domain. Beginning with a system-specific ODD definition, a set of core competencies are identified. These core competencies are then considered both in isolation and in conjunction with other potential confounding factors (e.g. other traffic or atmospheric conditions); with “edge cases” being represented as compounded or unique sets of confounding factors. Using this approach, a candidate scenario set is presented, along with a discussion of nuances and necessary considerations in scenario selection. These approaches are combined in a simulated environment to demonstrate their use. Finally, a strategy is proposed to automate the overall scenario testing process to make the execution less cumbersome. This process of test scenario creation strictly follows the ISO 26262 concept phase to verify the safety goals and functional safety requirements.
Patil, MayurLybarger, AlexanderMidlam-Mohler, ShawnStoddart, Evan
Multispeed eDrive or eAxles arguably benefit the overall performance and efficiency of an electric vehicle. The majority of the benefits can be derived from rightsizing and optimal control over the gear shift sequence under various driving scenarios. This paper focuses on developing an optimal shift schedule and precise shift controls for a special utility three-wheeled electric vehicle using a Model-Based Design approach. The supervisory control logic is implemented using Stateflow. Further, the entire shift mechanism with the controller, stepper motor and driver is modeled in the Matlab-Simulink-Simscape environment. A novel solution of integrating the SolidWorks CAD model of the gearbox provided by the manufacturer with the shift mechanism using Simulink Multibody is presented. Finally, the controller model and C- code test methods are presented to validate the behavior and functional requirements using MIL, SIL and PIL on a prototype microcontroller chip.
Dani, PriyankaMore, Raunak PraveenNegi, Vivek SinghDorle, Aniruddha
Model-based system simulations play a critical role in the development process of the automotive industry. They are highly instrumental in developing embedded control systems during conception, design, validation, and deployment stages. Whether for model-in-the-loop (MiL), software-in-the-loop (SiL) or hardware-in-the-loop (HiL) scenarios, high-fidelity plant models are particularly valuable for generating realistic simulation results that can parallel or substitute for costly and time-consuming vehicle field tests. In this paper, the development of a powertrain plant model and its correlation performance are presented. The focus is on the following modules of the propulsion systems: transmission, driveline, and vehicle. The physics and modeling approach of the modules is discussed, and the implementation is illustrated in Amesim software. The developed model shows good correlation performance against test data in dynamic events such as launch, tip-in, tip-out, and gearshifts. To quantify the correlation accuracy, metrics are defined to provide a numerical assessment of plant model behavior and accuracy. In this study the engine is treated as a torque source, to concentrate on transmission, driveline, and vehicle dynamics. If necessary, a full-fledged engine model based on GT-Power can be connected through co-simulation interface. Co-simulation of software from different vendors greatly expands the plant modeling capability and the flexibility of coupling with controllers. Co-simulated models for both SiL and HiL purposes are discussed, with the focus on the interface options, cosim running configurations, model reduction, and bench validation. Although the vehicle modeled here has a conventional powertrain with an automatic transmission, the modeling architecture and correlation methodology can be readily applied to alternative propulsion systems.
Zhou, JingKao, MinghuiCurran, Kristin
This paper presents an approach for performing software in the loop testing of autonomous vehicle software developed in the Autoware framework. Autoware is an open source software for autonomous driving that includes modules such as localization, detection, prediction, planning and control [8]. Multitudes of autonomous driving frameworks exist today, each having its own pros and cons. Often, MATLAB-Simulink is used for rapid prototyping, system modeling and testing, specifically for the lower-level vehicle dynamics and powertrain control features. For the autonomous software, the Robotic Operating System (ROS) is more commonly used for integrating distributed software components so that they can easily share information through a publish and subscribe paradigm. Thorough testing and evaluation of such complex, distributed software, implemented on a physical vehicle poses significant challenges in terms of safety, time, and cost, especially when considering rare edge cases. Virtual prototyping is therefore a crucial enabler in the development of autonomous software. In a simulated environment, many traffic scenarios under a variety of environmental conditions can be quickly evaluated, at low cost, without safety concerns. In this paper, we report on a simulation environment consisting of three simulation tools. PreScan (by Siemens/TASS) combined with Simulink (by Mathworks) is used for simulating how the vehicle interacts with the environment: sensors, actuators, the vehicle dynamics and powertrain. The autonomy software is emulated directly in Autoware.AI on top of ROS. To evaluate the autonomous software, synthetic data from the sensors simulated in PreScan are published to ROS where they are processed by the autonomy stack. Similarly, the control signals generated by the autonomy stack in Autoware are subscribed to by PreScan where they serve as input to the virtual vehicle model. The paper describes in detail the integration of PreScan and Autoware, illustrates this integration for object detection using a monocular camera, and characterizes the performance in terms of message transfer speed.
Bachuwar, SanketBulsara, ArdashirDossaji, HuzefaGopinath, AdityaParedis, ChrisPilla, SrikanthJia, Yunyi
Items per page:
1 – 50 of 134