Adaptive Nonlinear Model Predictive Cruise Controller: Trailer Tow Use Case

<div class="section abstract"><div class="htmlview paragraph">A variety of vehicle controls, from active safety systems to power management algorithms, can greatly benefit from accurate, reliable, and robust real-time estimates of vehicle mass and road grade. This paper develops a parallel mass and grade (PMG) estimation scheme and presents the results of a study investigating its accuracy and robustness in the presence of various noise factors. An estimate of road grade is calculated by comparing the acceleration as measured by an on-board longitudinal accelerometer with that obtained by differentiation of the undriven wheel speeds. Mass is independently estimated by means of a longitudinal dynamics model and a recursive least squares (RLS) algorithm using the longitudinal accelerometer to isolate grade effects. To account for the influences of acceleration-induced vehicle pitching on PMG estimation accuracy, a correction factor is developed from controlled tests under a wide range of throttle levels. The estimation approach is applied to data collected while driving on public roads under a variety of driving conditions, replicating several noise factors associated with daily driving. Thresholds are developed to isolate driving events that are likely to result in quick and accurate mass estimates, and an averaging filter is then applied to these converged values. The algorithm is shown to rapidly deliver estimates within 3% of true vehicle mass.</div></div>

<div class="section abstract"><div class="htmlview paragraph">For modern vehicle development, on-board vehicle Mass and road Gradient Estimation (MGE) can offer great benefit to many sub-systems on the vehicle, such as vehicle control system, transmission control system, and active safety system etc. However, there are still several challenges that need to be solved. Firstly, thanks to good accuracy, reliability, and robustness, regression analysis-based approaches: Recursive Least Squares (RLS) and Kalman Filter (KF) are very popular for MGE, but the trade-off between estimator’s accuracy and converge time is challenging. Furthermore, depending on vehicle and powertrain types, the implementation of MGE function could be very different. It is desired to have a structured approach for various vehicle applications’ MGE development. Lastly, good reliability of MGE does not always satisfy for complicated real-world driving maneuvers and road conditions.</div><div class="htmlview paragraph">This paper introduced a mature MGE development approach which utilizes an innovative Series Kalman Filter (S-KF) structure with a 3-axis accelerometer to address the challenges above, leading to the demonstration on 3 different types of vehicles with different powertrain applications: a truck with Internal Combustion Engine (ICE) and Automated Manual Transmission (AMT) gearbox, a mid-size Plug-in Hybrid Electric Vehicle (PHEV) SUV, and an A-segment Battery Electric Vehicles (BEV) mini car. They are tested under wide range real-world driving maneuvers and road conditions. Moreover, to improve the converge time whilst keeping the same accuracy of MGE, several dedicated designed engineering features are implemented and introduced in this paper. The vehicle-based experimental result and structured development approach showed good performance and maturity of the production oriented MGE, and good efficiency and flexibility to adapt it to various applications.</div></div>

<div class="section abstract"><div class="htmlview paragraph">A follow-up case study on rollover testing with a single full-size sport utility vehicle (SUV) was conducted under controlled real-world conditions. The purpose of this study was to conduct a well-documented rollover event that could be utilized in evaluating various methods and techniques over the phases associated with rollover accidents. The phases documented and discussed, inherent to rollovers, are: pre-trip, trip, and rolling phases.</div><div class="htmlview paragraph">With recent advances in technology, new devices and techniques have been designed which improve the ability to capture and document the unpredictable dynamic events surrounding vehicle rollovers. One such device is an inertial measurement unit (IMU), which utilizes GPS technology along with integrated sensors to report and record measured dynamic parameters real-time. The data obtained from a RT-4003 IMU device are presented and compared along with previous test data and methodology.</div><div class="htmlview paragraph">For this case study, an instrumented 1996 GMC Jimmy SUV was towed to 53.6 mph (86.3 kph) and released. An automated pre-programmed controller steered the vehicle through maneuvers intended to result in rollover. The SUV rolled 2 revolutions and was documented with both on and off-board instrumentation, videotape footage, survey measurements, and numerous high-resolution three-dimensional (3D) laser scans. The data were then compared and contrasted with previously published rollover techniques and methodologies.</div><div class="htmlview paragraph">This instrumented case study is a well-documented test that can be utilized in accident reconstruction for future evaluation, comparison, or simulation of similar type rollover events.</div></div>

<div class="htmlview paragraph">In this paper, we present a comprehensive approach to design of embedded real-time software for electrically controlled mechanical systems in automotive applications. As a case study, we implement a Gear change and Clutch controller for a Formula SAE car. This includes a generic communication interface and protocol for CAN bus communication, I/O interfaces for A/D conversion and PWM output, together with a PID controller for clutch actuation.</div> <div class="htmlview paragraph">Under our framework, the embedded software is developed using Timber, a programming language and formalism that provides executable models for embedded real-time systems.</div> <div class="htmlview paragraph">The case study shows how a complete control system can be straightforwardly modeled, simulated and transformed into executable code. The system has been realized and tested onto a lightweight, 8-bit AVR-5, embedded platform.</div> <div class="htmlview paragraph">Compared to the raw C code design flow, the proposed framework has in our case study showed increased efficiency with respect to development time. We boldly conclude that our Timber based framework offers true <b>“work with the work”.</b></div>

<div class="section abstract"><div class="htmlview paragraph">Future legislation scenarios as well as stringent CO<sub>2</sub> targets, in particular under real driving conditions, will require the introduction of new and additional powertrain technologies. Beside the increasing electrification of the powertrain, it will be essential to utilize the full potential of the Internal Combustion Engine (ICE). There is clearly a competition of new and different ICE-Technologies [<span class="xref">1</span>] including VCR. VCR systems are expected to be introduced to a considerable number of next generation turbocharged Spark Ignited (SI) engines in certain vehicle classes. The implementation of Miller or Atkinson cycles is an essential criterion for increased geometric Compression Ratio (CR). The DUAL MODE Variable Compression System (VCS)<sup>TM</sup> enables a 2-stage variation of the connecting rod length and thus of the compression ratio (CR). The concept allows a modular integration into existing engine families without significant modifications of the engine architecture and thus represents a cost-attractive solution. The basic functionality and potential were described in a technical paper which was presented at the SAE WCX 2017 [<span class="xref">2</span>], the development of the hydraulic function and oil investigations were presented at the SAE WCX 2018 [<span class="xref">3</span>].</div><div class="htmlview paragraph">This paper describes the engine integration of the Dual Mode VCS-system including the hydraulic control system into a state-of-the-art 2.0L-Turbo Gasoline Direct Injected (TGDI)-Engine for the vehicle boundaries of a midsize Sports Utility Vehicle (SUV).</div><div class="htmlview paragraph">Switching strategies, as well as the importance of short switching duration under Worldwide Harmonized Light Vehicle Test Procedure (WLTP) as well as Real Driving Emissions (RDE) requirements are discussed.</div><div class="htmlview paragraph">The applied fail-safe operation and Dual Mode VCS position control and detection will be described.</div><div class="htmlview paragraph">Finally, the Fuel Consumption (FC) impact of the Dual Mode VCS system is discussed. For the 2.0L-TGDI engine in the actual vehicle class, dependent on the load profile and the specific power and torque requirements a CO<sub>2</sub> reduction between 5% and up to 9% in the WLTP is achievable.</div></div>

<div class="section abstract"><div class="htmlview paragraph">Traditionally, ground vehicle aerodynamics has been researched with highly simplified models such as the Ahmed body and the SAE model. These models established and advanced the fundamental understanding of bluff body aerodynamics and have generated a large body of published data, however, their application to the development of passenger vehicles is limited by the highly idealized nature of their geometries. To date, limited data has been openly published on aerodynamic investigations of production vehicles, most likely due to the proprietary nature of production vehicle geometry. In 2012, Heft et al. introduced the realistic generic car model ‘DrivAer’ that better represents the flow physics associated with a typical production vehicle. The introduction of the DrivAer model has led to a broad set of published data for both experimental and computational investigations and has proven itself invaluable as a correlation and calibration tool of wind tunnels, the validation of computational fluid dynamics (CFD) codes and increasing the understanding of the fundamental flow physics around passenger vehicles.</div><div class="htmlview paragraph">Automotive sales trends in the United States, published by the Bureau of Economic Analysis in 2018, indicate that sales of Pickup Trucks (PUs) and Sports Utility Vehicles (SUVs) have increased over the past 10 years and are outselling sedans at a rate just over two to one. Compared to sedans, PU and SUV body styles pose additional aerodynamic challenges due to their complex wake structures. The introduction of a realistic generic PU and SUV model as an open access tool is expected to yield benefits to the wider community, equivalent to those of the DrivAer passenger vehicle. This paper introduces the Generic Truck Utility (GTU) as a realistic, generic PU truck and interchangeable SUV model. The paper will focus on the design and development of the GTU and will present a summary of preliminary experimental results of the GTU complemented by numerical simulations using iconCFD®, an open source based CFD solver.</div></div>

<div class="section abstract"><div class="htmlview paragraph">A methodology to design a model free cruise control algorithm(MFCC) is presented in this paper. General cruise control algorithms require lots of vehicle parameters to control the power train and the brake system, that makes control system complicate. Moreover, when the target vehicle is changed, the vehicle parameters should be reinvestigated in order to apply the cruise control algorithm to the subject vehicle. To overcome these disadvantages of the conventional cruise control algorithm, MFCC algorithm has been developed. The algorithm directly determines the throttle, brake inputs based on the reference model parameters such as clearance, relative velocity, and subject vehicle acceleration. This simple structure facilitates human centered design of cruise controller and makes it easy to apply control algorithm to various vehicles without reinvestigation of vehicle parameters. To achieve vehicle safety and driver comfort, control parameters of the model free cruise control algorithm has been designed using the real-world driving test data. Few parameters have been adapted using a learning algorithm to minimize the effects of unawareness of the vehicle parameters. From the simulation study, the performance of the proposed controller has been compared with that of the conventional adaptive cruise controller(ACC). It is shown that the proposed control strategy its behavior is not only similar to that of the conventional ACC in normal driving condition but also more robust than the ACC for model uncertainty and external disturbance.</div></div>

<div class="section abstract"><div class="htmlview paragraph">Vehicle testing often requires accurate speed control, whether maintaining a constant speed or following a dynamic speed profile. A portable brake and throttle robot designed for this task must quickly and automatically adapt its control to each test vehicle's mass and powertrain characteristics in order for the controller to perform well on a wide variety of platforms.</div><div class="htmlview paragraph">This application presents two major challenges. First, each new vehicle installation requires rapid retuning of the controller. Manual tuning can be very time-consuming. Second, the “plant” is very nonlinear and asymmetric. Different actuators are employed for acceleration (engine) and braking (road loads and friction brakes).</div><div class="htmlview paragraph">After a broad survey of control strategies and their suitability for this particular application, neurofuzzy techniques were among the most promising. Neurofuzzy networks can not only approximate nonlinear functions accurately, but the fuzzy rule-consequent weights can be readily updated in real time.</div><div class="htmlview paragraph">A neurofuzzy controller was designed for the automotive speed tracking problem. Experimental testing was conducted and results were presented and analyzed.</div></div>

<div class="section abstract"><div class="htmlview paragraph">This study presents a novel approach for predicting fuel consumption in heavy-duty vehicles using a Machine Learning-based model, which is based on feedforward neural network (FFNN). The model is designed to enhance real-time vehicle monitoring, optimize route planning, and reduce both operational costs and environmental impact, making it particularly suitable for fleet management applications. Unlike traditional physics-based approaches, the FFNN relies solely on a refined selection of input variables, including vehicle speed, acceleration, altitude, road slope, ambient temperature, and engine power. Additionally, vehicle mass is estimated using a methodology presented elsewhere and is included as an input for a better generalization of the consumption model. This parameter significantly impacts fuel consumption and is particularly challenging to obtain for heavy-duty vehicles. Engine power is derived from both engine torque and speed (RPM), ensuring a direct relationship with fuel consumption while keeping computational complexity low.</div><div class="htmlview paragraph">Experimental data were collected from a fleet of heavy-duty trucks under real-world operating conditions. In particular, the study focuses on a turbocharged diesel truck with a maximum power output of 353 kW. The acquisition system is based on an On-Board Unit (OBU) featuring CANbus connectivity, GPS tracking and 4G/LTE-5G communication. The OBU enables continuous logging of vehicle and engine parameters over each mission. Additional road data, such as altitude and slope, were obtained from external mapping services.</div><div class="htmlview paragraph">For the purpose of FFNN development, the training dataset was compiled from approximately 10 different routes, capturing diverse driving conditions, while validation was conducted on 10 independent routes to assess model generalization. Before training, input data were pre-processed using normalization and standardization techniques to ensure stable convergence and mitigate the impact of scale differences among input variables. A correlation analysis was performed to evaluate the relationships among available parameters and fuel consumption, guiding the selection of the most informative inputs. This step reduced redundancy in the dataset and improved network efficiency. Furthermore, hyperparameter optimization was conducted using a Randomized Grid Search algorithm, enabling the identification of an optimal network architecture - specifically in terms of the number of layers and hidden neurons - and training parameters while minimizing overfitting.</div><div class="htmlview paragraph">The final model demonstrated high predictive accuracy across various validation routes, confirming the effectiveness of the FFNN in estimating fuel consumption with a reduced input set. Accuracy was tested on up to 50 routes, whose data where not considered for both training and validation; it was assessed that fuel consumption percentage error per route never exceeded 2%. This approach provides a practical and computationally efficient solution for fleet operators, facilitating advanced route planning and enabling more sustainable transportation strategies through cost-effective fuel management and emissions reduction.</div></div>

<div class="section abstract"><div class="htmlview paragraph">Sports Utility Vehicles (SUVs) Braking performance is a critical parameter, and it is described by the measure of stopping distance & stability of the vehicle at a specified speed. When it comes to electric vehicles, balancing tyre performance parameters like RRc, traction, wear, ride and handling, weight and NVH is challenging due to higher axle weights, higher moment of inertia and high instant torque delivery. This study presents the braking performances of EV-intended tyres at various contact surfaces in both dry & wet conditions. Impact of tyre design attributes like tyre tread pattern, contact patch, tread compound, tyre carcass construction & characteristics studied on 3 different tyres including ultra-high-performance tyres on braking performances. This paper contains the study & analysis of 1) Footprint analysis concerning the shape of the contact patch, pressure distribution, land-sea ratio, and contact area 2)Tyre tread design such as NSD, grooves, ribs & sipes design and pitch length 3) Compound property studies like hardness, modulus, tensile strength, tear strength & elongation at break 4)Materials like polymer blend, filler ratio & new-gen materials 5)Tyre performance characteristics which are contributing to vehicle stability like lateral stiffness, longitudinal stiffness, dynamic stiffness & mu slip curves and 6) Longitudinal & lateral forces at various speed conditions with CAE Simulations.</div><div class="htmlview paragraph">BEV requires low rolling resistance for improving vehicle range and which calls for reducing the hysteresis of tyre tread compound which affect vehicle barking performance at wet surface and it should be compensated by adjusting other design parameters.</div><div class="htmlview paragraph">Mu slip curve very important characteristics for designing brakes system to get peak mu and programming software for ABS system.</div></div>

<div class="htmlview paragraph">Tomorrow's winning powertrain solutions reside in those technology combinations providing optimized propulsion systems with zero emissions and no cost or performance penalty compared with today's vehicles.</div> <div class="htmlview paragraph">The recent Kyoto Protocol for CO<sub>2</sub> reduction and the California Air Resources Board (CARB) thrust for zero emission vehicles along with the European Regulatory community, motivate car manufacturers to adopt new light body structures with low aerodynamic drag coefficients, low-rolling resistance and the highest efficiency powertrains. The environmental equation expresses car manufacturers aptitude and desire to create zero emission vehicles at acceptable levels of performance unlike limited range electrical powered vehicle products.</div> <div class="htmlview paragraph">The cheapest solution to the environmental equation remains the conventional internal combustion engine ($30 to $50 per kW). However, new system optimization is necessary and will take many forms from alternative fuels enriched with hydrogen to diesel direct injected engines for light duty trucks and sport utility vehicles. Diesel fuel will probably be the favorite mid-term option with direct injection engines coupled with an electric motor. This combination gives higher efficiency, but a new optimized aftertreatment system will be necessary to satisfy emission mandates.</div> <div class="htmlview paragraph">Hybrid vehicles, the Toyota Prius being an excellent example, will increase in popularity with their high efficiency and driving smoothness.</div>

<div class="section abstract"><div class="htmlview paragraph">The longitudinal dynamics control is an essential task of vehicle dynamics control. In present, it is usually applied by adjusting the slip ratio of driving wheels to achieve satisfactory performances both in stability and accelerating ability. In order to improve its performances, the coordination of different subsystems such as engine, transmission and braking system has to be considered. In addition, the proposed algorithms usually adopt the threshold methods based on less road condition information for simpleness and quick response, which cannot achieve optimal performance on various road conditions.</div><div class="htmlview paragraph">In this paper, an integrated longitudinal vehicle dynamics control algorithm with tire/road friction estimation was proposed. First, a road identification algorithm was designed to estimate tire forces of driving wheels and the friction coefficient by the method of Kalman Filter and Recursive Least Squares (RLS). Then, a rule based integrated control algorithm which coordinate the engine torque control, brake pressure control and transmission shifting control was built to improve the vehicle driving performance.</div><div class="htmlview paragraph">During the longitudinal dynamics control procedure, the transmission shifting control algorithm firstly decided whether the vehicle starting up at the 1st or the 2nd gear based on the road condition. Then, the engine torque control algorithm was applied to adjust the slip ratio of driving wheels. Its process was divided into three phases. In each phase, different control rules with integration of engine torque and brake torque were set by the combined feed forward and feedback methods to meet with requirements of accuracy, rapidity and stability. The brake pressure was determined by method of sliding mode control (SMC) for its rapid adjustment characteristic, in the meanwhile, it was also an important component to be considered in engine torque feedback control.</div><div class="htmlview paragraph">Finally, the algorithm was verified by using Matlab/Simulink and CarSim co-simulation, the results show that the proposed control algorithm could regulate slip ratios of driving wheels fast and accurately, and the vehicle driving performance could be improved effectively.</div></div>

<div class="htmlview paragraph">Radar based collision warning systems (CWS) for heavy trucks have been in production and on the road in the United States for approximately two years. A short description of these systems is presented with actual driver performance data and accident reduction data as experienced by various U.S. national fleets using these systems. The next application of radar sensors is for Intelligent Cruise Control (ICC) or also known as Adaptive Cruise Control (ACC). This paper describes the operation of adaptive cruise control and presents the associated benefits to the driver</div> <div class="htmlview paragraph">The radar technology has been enhanced and the application has been extended to include radar controlled adaptive cruise control for heavy trucks This has been made possible by 1) technology improvements in radar systems, plus 2) the addition of the standard SAE J1939 data bus for engine controllers on most U.S diesel engines. The SAE J1939 data bus allows electronic control of engine speed by the radar system plus in many cases, control of engine retarder braking to assist in vehicle slowing while in adaptive cruise control operation.</div> <div class="htmlview paragraph">This paper described the radar operation, radar enhancements and the benefits of adaptive cruise control as experienced by drivers. The most significant radar enhancement, called monopulse radar is presented Comparisons are made between standard cruise control operation on heavy trucks and adaptive cruise control</div>

<div class="htmlview paragraph">Predictive Cruise Control (PCC) is a system that enhances and works in combination with the existing Conventional Cruise Control. Based on elevation information captured in a 3D map and a predictive algorithm, PCC allows the vehicle speed to vary around the cruise control set speed within a defined speed band in an effort to reduce fuel consumption. As fuel consumption is a major portion of a truck's life cycle costs (LCC) and cruise control is used extensively in the United States and Canada, PCC can significantly reduce the truck's LCC.</div>

<div class="section abstract"><div class="htmlview paragraph">One often reported roadblock to consumer acceptance of electric vehicles is driving range, which is a function of powertrain efficiency and vehicle mass. Electric vehicle gearbox design often is based on multiple parallel shafts, thereby creating significant packaging constraints. Industry perception holds that deep groove ball bearings (DGBB) are more efficient than tapered roller bearings (TRB), and standard spin-loss testing confirms those beliefs. However, spin-loss efficiency testing does not accurately reproduce typical real-world driving. A more realistic comparison of bearing efficiency is required to properly select bearings during the powertrain design stage.</div><div class="htmlview paragraph">Recently completed testing focused on recreating application conditions (including bearing loads, speeds, misalignment, and load zones) for electric vehicle gearbox intermediate shafts. These conditions varied between TRB and DGBB as a result of application bearing setting and reaction load changes and were faithfully reproduced. Under light load conditions, power loss from a 35 mm DGBB was significantly lower than that of a 35 mm bore TRB, but the TRB power loss was significantly lower than the DGBB’s under heavier loads.</div><div class="htmlview paragraph">When the individual test conditions were weighted to represent a road-use cycle, weighted power loss was equivalent between the two bearing types at expected vehicle operating temperatures. The bearings had the same rated radial capacity, but the TRB had reduced width, outer diameter (OD), and mass. The TRB’s capability of exhibiting power loss equivalent to the DGBB but within a smaller required envelope size can significantly aid electric vehicle design engineers who are faced with fitting multiple parallel shafts in their applications. In addition, the reduced size of the TRB allows system mass reductions, increasing battery range to address consumer concerns.</div></div>