Search Results

Abstract Autonomous vehicles (AVs) rely heavily upon their perception subsystems to “see” the environment in which they operate. Unfortunately, the effect of variable weather conditions presents a significant challenge to object detection algorithms, and thus, it is imperative to test the vehicle extensively in all conditions which it may experience. However, the development of robust AV subsystems requires repeatable, controlled testing—while real weather is unpredictable and cannot be scheduled. Real-world testing in adverse conditions is an expensive and time-consuming task, often requiring access to specialist facilities. Simulation is commonly relied upon as a substitute, with increasingly visually realistic representations of the real world being developed.

Abstract The key issues of automatic emergency braking (AEB) control algorithm are when and how to brake. This article proposes an AEB control algorithm that integrates risk perception (RP) and emergency braking characteristics of professional drivers for rear-end collision avoidance. Using the formulated RP by time to collision (TTC) and time headway (THW), the brake trigger time can be determined. Based on the professional driver fitting (PDF) characteristic, the brake pattern can be developed. Through MATLAB/Simulink simulation platform, the European New Car Assessment Programme (Euro-NCAP) test scenarios are used to verify the proposed control algorithm. The simulation results show that compared with the TTC control algorithm, PDF control algorithm, and the integrated PDF and TTC control algorithm, the proposed integrated PDF and RP control algorithm has the best performance, which can not only ensure safety and brake comfort, but also improve the road resource utilization rate.

Abstract Automated driving as one of the most anticipated technologies is approaching its market release in the near future. Since several years, the research in the automotive industry is largely focused on its development and presents well-engineered prototypes. The many aspects of this development do not only concern the function and its components itself, but also the proof of safety and assessment for its market release. It is clear that previous methods used for the release of Advanced Driver Assistance Systems are not applicable. In contrast to already released systems, automated driving is not restricted to a certain field of application in terms of driving scenarios it has to take action in. This results in an infeasible amount of required testing and unforeseeable scenarios the function can face throughout its lifetime. In this article, we show a scenario-based approach that promises to overcome those challenges.

Abstract A tandem axle suspension is an important system to the ride comfort and vehicle stability of and road damage experience from commercial vehicles. This article introduces an investigation into the use of a controlled active tandem axle suspension, which for the first time enables more effective control using two fuzzy logic controllers (FLC). The proposed controllers compute the actuator forces based on system outputs: displacements, velocities, and accelerations of movable parts of tandem axle suspension as inputs to the controllers, in order to achieve better ride comfort and vehicle stability and extend the lifetime of road surface than the conventional passive suspension. A mathematical model of a six-degree-of-freedom (6-DOF) tandem axle suspension system is derived and simulated using Matlab/Simulink software.

Abstract In this article, a hierarchical coordinated control algorithm for integrating active rear steering and driving/braking force distribution (ARS+D/BFD) was presented. The upper-level control was synthesized to generate the required rear steering angle and external yaw moment by using a sliding-mode controller. In the lower-level controller, a control allocation algorithm considering driving/braking actuators and tire forces constraints was designed to assign the desired yaw moment to the four wheels. To this end, an optimization problem including several equality and inequality constraints were defined and solved analytically. Finally, computer simulation results suggest that the proposed hierarchical control scheme was able to help to achieve substantial enhancements in handling performance and stability.

Abstract In this study, a computer-aided design (CAD) geometry system that is linked to each other to create a parametric form of the side rear door’s inner frame sheet piece on a passenger vehicle body in a Siemens NX environment was developed. The system was created in the NX CAD environment, using the program’s unique product development structure. The system was designed and modified for time-consuming parts. At the end of the study, the parameterized vehicle door geometries worked in the NX environment standardized the design process and accelerated the design works.

Abstract Throughout the automobile industry, the electronic brake boost technologies have been widely applied to support the expansion of the using range of the driver assist technologies. The electronic brake booster (EBB) supports to precisely operate the brakes as necessary via building up the brake pressure faster than the vacuum brake booster. Therefore, in this article a novel control strategy for the EBB based on fuzzy logic control (FLC) is developed and studied. The configuration of the EBB is established and the system model including the permanent magnet synchronous motor (PMSM), a two-stage reduction transmission (gears and a ball screw), a servo body, reaction disk, and the hydraulic load are modeled by MATLAB/Simulink. The load-dependent friction has been compensated by using Karnopp friction model. Due to the strong nonlinearity on the EBB components and the load-dependent friction, FLC has been used for the control algorithm.

Abstract As automated driving technology is evolving quickly and becomes more widely deployed, it is essential to validate the Safety of the Intended Functionality (SOTIF) of Automated Vehicles (AVs) prior to mass production. In general, an exhaustive real-world scenario validation of AVs is considered infeasible due to excessive time consumption. Additionally, simulation tests alone are often regarded as inadequate since it is difficult to model the system and physical properties of vehicles with full fidelity. Therefore, a SOTIF validation method for AVs is proposed in this article, which consists of structure design and scenario determination. A mature, systematic, and complete set of testing and evaluation procedures is presented in structure design, and a scenario generation method is introduced in scenario determination. The SOTIF validation method takes advantage of both simulation tests and on-road tests.

Abstract Constant area section length downstream to the fuel injection point is a crucial dimension of scramjet duct geometry. It has a major contribution in creating the maximum effective pressure inside the combustor that is required for propulsion. The length is limited by the thermal choking phenomenon, which occurs when heat is added in a flow through constant area duct. As per theory, to avoid thermal choking the constant area section length depends upon the inlet conditions and the rate of heat addition. The complexity related to mixing and combustion process inside the supersonic stream makes it difficult to predict the rate of heat addition and in turn the length. Recent efforts of simulating the reacting flow inside scramjet combustors are encouraging and can be useful in this regard. The presented work attempts to use simulation results of scramjet combustion for predicting the constant area section length for a typical scramjet combustor.

Abstract The automotive industry offers many applications for machine learning (ML), in general, and deep neural networks in particular. However, the real-world deployment of neural networks into safety-critical components remains a challenge as models would need to offer robustness under a wide range of operating conditions. In this work, we focus on uncertainty estimation, which can be used to deliver predictors that fail gracefully, by detecting situations where their predictions are unreliable. Following Gräber et al. [1], we use Recurrent Neural Networks (RNNs) to perform sideslip angle estimation. To perform robust uncertainty estimation, we augment the RNNs with generative models. We demonstrate the advantage of the proposed model architecture over Monte Carlo (MC) dropout [2] on the Revs data set [3].

Abstract To reduce fuel consumption and carbon dioxide (CO2) emissions from mobile air conditioning (A/C) systems, “U.S. Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards” identified solar/thermal technologies such as solar control glazings, solar reflective paint, and active and passive cabin ventilation in an off-cycle credit menu. National Renewable Energy Laboratory (NREL) researchers developed a sophisticated analysis process to calculate U.S. light-duty A/C fuel use that was used to assess the impact of these technologies, leveraging thermal and vehicle simulation analysis tools developed under previous U.S. Department of Energy projects. Representative U.S. light-duty driving behaviors and weighting factors including time-of-day of travel, trip duration, and time between trips were characterized and integrated into the analysis.

Abstract Low-Pressure cooled Exhaust Gas Recirculation (LP-cEGR) is proven to be an effective technology for fuel efficiency improvement in turbocharged spark-ignition (SI) engines. Aiming to fully exploit the EGR benefits, new challenges are introduced that require more complex and robust control systems and strategies. One of the most important restrictions of LP-cEGR is the transient response, since long air-EGR flow paths introduce significant transport delays between the EGR valve and the cylinders. High dilution generally increases efficiency, but can lead to cycle-by-cycle combustion variation. Especially in SI engines, higher-than-requested EGR dilution may lead to combustion instabilities and misfires. Considering the long EGR evacuation period, one of the most challenging transient events is throttle tip-out, where the engine operation shifts from a high-load point with high dilution tolerance to a low-load point where EGR tolerance is significantly reduced.

Abstract The external driving environment of an autonomous driving vehicle is complex and changeable. In this article, the trajectory tracking control with obstacle avoidance based on model predictive control was presented. Specifically, double-level control scheme by controlling the front steering angle was used in our research, and the double level is composed of the high level of model predictive controller for local trajectory planning and low level of model predictive controller for trajectory tracking. At high level, the local trajectory planner based on the point-mass model was designed. Then, at low level, the linear time-varying vehicle dynamics model was presented, and the trajectory tracking controller was proposed considering control variable, control increment, and output constraint. Finally, the trajectory tracking performance was tested in co-simulation environment with CarSim and Simulink, and the tracking errors were analyzed.

Abstract The promise and potential for a future of automated vehicles (AVs) remains great, with safety and societal transformations that may rival the original introduction of the automobile. Yet an inability for industry and governments to define what it means for an AV to drive safely has tempered enthusiasm and risks causing a “winter of AV” just like the one that affected Artificial Intelligence technologies decades ago, which is only now being overcome. Towards this end, the Responsibility-Sensitive Safety (RSS) model was introduced as an open and transparent white-box, an interpretable and scalable formal model that defines minimum safety requirements based on reasonable assumptions of others, balancing safety and usefulness for automated driving vehicles.

Abstract Emerging technologies for connected and automated vehicles (CAVs) are rapidly advancing, and there is an incremental adoption of partial automation systems in existing vehicles. Nevertheless, there are still significant barriers before fully or highly automated vehicles can enter mass production and appear on public roads. These are not only associated with the need to ensure their safe and efficient operation but also with cost and delivery time constraints. A key challenge lies in the testing and validation (T&V) requirements of CAVs, which are expected to be significantly higher than those of traditional and partially automated vehicles. Promising methodologies that can be used toward this goal are scenario-based (SBT) and X-in-the-Loop (XiL) testing. At the same time, complex techniques such as co-simulation and mixed-reality simulation could also provide significant benefits.

Abstract This article presents a new machine learning (ML) development lifecycle which will constitute the core of the new aeronautical standard on ML called AS6983, jointly being developed by working group WG-114/G34 of EUROCAE and SAE. The article also presents a survey of several existing standards and guidelines related to ML in aeronautics, automotive, and industrial domains by comparing and contrasting their scope, purpose, and results.

Abstract Scenario-based testing is a promising approach to solving the challenge of proving the safe behavior of vehicles equipped with automated driving systems (ADS). Since an infinite number of concrete scenarios can theoretically occur in real-world road traffic, the extraction of scenarios relevant in terms of the safety-related behavior of these systems is a key aspect for their successful verification and validation. Therefore, a method for extracting multimodal urban traffic scenarios from naturalistic road traffic data in an unsupervised manner, minimizing the amount of (potentially biased) prior expert knowledge, is proposed. Rather than an (elaborate) rule-based assignment by extracting concrete scenarios into predefined functional scenarios, the presented method deploys an unsupervised machine learning pipeline. The approach allows for exploring the unknown nature of the data and their interpretation as test scenarios that experts could not have anticipated.

Abstract A torque converter was instrumented with 29 pressure transducers inside five cavities under study (impeller, turbine, stator, clutch cavity between the pressure plate and the turbine shell). A computer model was created to establish correlation with measured torque and pressure. Torque errors between test and simulation were within 5% and K-Factor and torque ratio errors within 2%. Turbulence intensity on the computer model was used to simulate test conditions representing transmission low and high line pressure settings. When turbulence intensity was set to 5%, pressure simulation root mean square errors were within 11%-15% for the high line pressure setting and up to 34% for low line pressure setting. When turbulence intensity was increased to 50% for the low line pressure settings, a 6% reduced root mean square error in the pressure simulations was seen.

Abstract Non-pneumatic tires (NPTs) have been widely used due to their advantages of no occurrence of puncture-related problems, no need of air maintenance, low rolling resistance, and improvement of passenger comfort due to its better shock absorption. It has a variety of applications as in earthmovers, planetary rover, stair-climbing vehicles, and the like. Recently, the unique puncture-proof tire system (UPTIS) NPT has been introduced for passenger vehicles segment. The spoke design of NPT-UPTIS has a significant effect on the overall working performance of tire. Optimized tire performance is a crucial factor for consumers and original equipment manufacturers (OEMs). Hence to optimize the spoke design of NPT-UPTIS spoke, the top and bottom curve of spoke profile have been described in the form of analytical equations. A generative design concept has been introduced to create around 50,000 spoke profiles.

Abstract Threat identification and security analysis have become mandatory steps in the engineering design process of high-assurance systems, where successful cyberattacks can lead to hazardous property damage or loss of lives. This article describes a novel approach to perform security analysis on embedded systems modeled at the architectural level. The tool, called Security Threat Evaluation and Mitigation (STEM), associates threats from the Common Attack Pattern Enumeration and Classification (CAPEC) library with components and connections and suggests potential defense patterns from the National Institute of Standards and Technology (NIST) Special Publication (SP) 800-53 security standard. This article also provides an illustrative example based on a drone package delivery system modeled in AADL.