Search Results

The SAE AutoDrive Challenge II is a four-year collegiate competition dedicated to developing a Level 4 autonomous vehicle by 2025. In January 2023, the participating teams each received a Chevy Bolt EUV. Within a span of five months, the second phase of the competition took place in Ann Arbor, MI. The authors of this contribution, who participated in this event as team Wisconsin Autonomous representing the University of Wisconsin–Madison, secured second place in static events and third place in dynamic events. This has been accomplished by reducing reliance on the actual vehicle platform and instead leveraging physical analogs and simulation. This paper outlines the software and hardware infrastructure of the competing vehicle, touching on issues pertaining sensors, hardware, and the software architecture employed on the autonomous vehicle. We discuss the LiDAR-camera fusion approach for object detection and the three-tier route planning and following systems.

Lithium-ion and Lithium polymer batteries are fast becoming ubiquitous in high-discharge rate applications for military and non-military systems. Applications such as small aerial vehicles and energy transfer systems can often function at C-rates greater than 1. To maximize system endurance and battery health, there is a need for models capable of precisely estimating the battery state-of-charge (SoC) under all temperature and loading conditions. However, the ability to perform state estimation consistently and accurately to within 1% error has remained unsolved. Doing so can offer enhanced endurance, safety, reliability, and planning, and additionally, simplify energy management. Therefore, the work presented in this paper aims to study and develop experimentally validated mathematical models capable of high-accuracy battery SoC estimation.

Identifying edge cases for designed algorithms is critical for functional safety in autonomous driving deployment. In order to find the feasible boundary of designed algorithms, simulations are heavily used. However, simulations for autonomous driving validation are expensive due to the requirement of visual rendering, physical simulation, and AI agents. In this case, common sampling techniques, such as Monte Carlo Sampling, become computationally expensive due to their sample inefficiency. To improve sample efficiency and minimize the number of simulations, we propose a tailored active learning approach combining the Support Vector Machine (SVM) and the Gaussian Process Regressor (GPR). The SVM learns the feasible boundary iteratively with a new sampling point via active learning. Active Learning is achieved by using the information of the decision boundary of the current SVM and the uncertainty metric calculated by the GPR.

5 belt wind tunnels are the most common facility to conduct the experimental aerodynamics development for production cars. Among aerodynamic properties, usually drag is the most important development target, but lift force and its front/rear balance is also important for vehicle dynamics. Related to the lift measurement, it is known that the “pad correction”, the correction in the lift measurement values for the undesirable aerodynamic force acting on wheel belt surface around the tire contact patch, must be accounted. Due to the pad correction measurement difficulties, it is common to simply subtract a fixed amount of lift values from measured lift force. However, this method is obviously not perfect as the pad corrections are different for differing vehicle body shapes, aerodynamic configurations, tire sizes and shapes.

Traffic state identification is a key problem in intelligent transportation system. As a new technology, connected and automated vehicle can play a role of identifying traffic state with the installation of onboard sensors. However, research of lane level traffic state identification is relatively lacked. Identifying lane level traffic state is helpful to lane selection in the process of driving and trajectory planning. In addition, traffic state identification precision with low penetration of connected and automated vehicles is relatively low. To fill this gap, this paper proposes a novel method of identifying traffic state in the presence of connected and automated vehicles with low penetration rate. Assuming connected and automated vehicles can obtain information of surrounding vehicles’, we use the perceptible information to estimate imperceptible information, then traffic state of road section can be inferred.

Appropriate spray modeling in multidimensional simulations of diesel engines is well known to affect the overall accuracy of the results. More and more accurate models are being developed to deal with drop dynamics, breakup, collisions, and vaporization/multiphase processes; the latter ones being the most computationally demanding. In fact, in parallel calculations, the droplets occupy a physical region of the in-cylinder domain, which is generally very different than the topology-driven finite-volume mesh decomposition. This makes the CPU decomposition of the spray cloud severely uneven when many CPUs are employed, yielding poor parallel performance of the spray computation. Furthermore, mesh-independent models such as collision calculations require checking of each possible droplet pair, which leads to a practically intractable O(np2/2) computational cost, np being the total number of droplets in the spray cloud, and additional overhead for parallel communications.

We present an approach in which an open-source software infrastructure is used for testing the behavior of autonomous vehicles through computer simulation. This software infrastructure is called CAVE, from Connected Autonomous Vehicle Emulator. As a software platform that allows rapid, low-cost and risk-free testing of novel designs, methods and software components, CAVE accelerates and democratizes research and development activities in the field of autonomous navigation.

Currently, car bodies require further weight reduction in order to support increasing fuel economy requirements. An efficient way for light weight body design is to include body member size as a design variable in addition to part thickness. However it is currently difficult for finite element (FE) models to change member size even using current morphing techniques. To break through this challenge, a hybrid modeling approach was developed which combines shell and beam element representations of body structural members. The original member shell element thickness was decreased by 40%. Then the stiffness reduction caused by this change is offset by beam elements incorporated inside these members. These beams can represent the stiffness change due to new cross sectional dimensions or orientations without changing the original shell elements, thus avoiding modeling instabilities that can occur from morphing.

In recent years, the increased use of electric power steering in vehicles has increased the importance of issues such as making systems more compact and lightweight, and dealing with increased development man-hours. To increase development efficiency, the use of a “Hardware in the loop simulator” (HILS) is being tested to shift from the previous development method that relied on a driver's subjective evaluation in an actual vehicle test to bench-test development. Using HILS enables tasks such as specification studies, performance forecasts, issue identification and countermeasure proposals to be performed at an early stage of development even when there is no prototype vehicle. This report describes a case study of using HILS to solve the issues of reducing the load by adjusting the geometric specifications around the kingpin and eliminating the tradeoff by adding a new EPS control algorithm in order to make the electric power steering (EPS) more compact and lightweight.

Optimal fuel injection strategies are obtained with a micro-genetic algorithm and an adaptive gradient method for a nonroad, medium-speed DI diesel engine equipped with a multi-orifice, asynchronous fuel injection system. The gradient optimization utilizes a fast-converging backtracking algorithm and an adaptive cost function which is based on the penalty method, where the penalty coefficient is increased after every line search. The micro-genetic algorithm uses parameter combinations of the best two individuals in each generation until a local convergence is achieved, and then generates a random population to continue the global search. The optimizations have been performed for a two pulse fuel injection strategy where the optimization parameters are the injection timings and the nozzle orifice diameters.

A global optimization method has been developed for an engine simulation code and utilized in the search of optimal fuel injection strategies. This method uses a Lagrange interpolation function which interpolates engine output data generated at the vertices and the intermediate points of the input parameters. This interpolation function is then used to find a global minimum over the entire parameter set, which in turn becomes the starting point of a CFD-based optimization. The CFD optimization is based on a steepest descent method with an adaptive cost function, where the line searches are performed with a fast-converging backtracking algorithm. The adaptive cost function is based on the penalty method, where the penalty coefficient is increased after every line search. The parameter space is normalized and, thus, the optimization occurs over the unit cube in higher-dimensional space.

Combustion simulation is an effective tool in overcoming the issues associated with gasoline HCCI engines, controlling ignition timing and extending the operating range. The research discussed in this paper commenced by optimizing the reaction mechanism from the perspective of ignition delay using the genetic algorithm (GA) method. Simulations employing the optimized reaction mechanism were then able to more accurately reproduce the ignition timing of iso-octane and primary reference fuels (PRF). Ignition times obtained from simulations showed excellent correlation with ignition times measured using these fuels in shock tube experiments, and in engines with both homogeneous and non-homogeneous fuel distributions. The use of the PRF mechanism for gasoline with an equivalent octane number enables excellent reproduction of ignition timing even when EGR is employed.

An accurate and rapid engine speed control has been required to improve the shift quality of automated manual transmissions. However, conventional controls could not provide sufficient controllability. They caused the overshoot and steady-state error of engine speed from target values. Therefore, a two-degree-of-freedom sliding mode algorithm was newly designed and applied to the engine speed control. This algorithm can independently assign the disturbance suppression characteristic and tracking performance, and has excellent robustness against the changes in engine dynamics. As a result, the overshoot and steady-state error were prevented under all engine conditions.

Neural networks are useful tools for optimization studies since they are very fast, so that while capturing the accuracy of multi-dimensional CFD calculations or experimental data, they can be run numerous times as required by many optimization techniques. This paper describes how a set of neural networks trained on a multi-dimensional CFD code to predict pressure, temperature, heat flux, torque and emissions, have been used by a genetic algorithm in combination with a hill-climbing type algorithm to optimize operating parameters of a diesel engine over the entire speed-torque map of the engine. The optimized parameters are mass of fuel injected per cycle, shape of the injection profile for dual split injection, start of injection, EGR level and boost pressure. These have been optimized for minimum emissions. Another set of neural networks have been trained to predict the optimized parameters, based on the speed-torque point of the engine.

Neural networks have been used for engine computations in the recent past. One reason for using neural networks is to capture the accuracy of multi-dimensional CFD calculations or experimental data while saving computational time, so that system simulations can be performed within a reasonable time frame. This paper describes three methods to improve upon neural network predictions. Improvement is demonstrated for in-cylinder pressure predictions in particular. The first method incorporates a physical combustion model within the transfer function of the neural network, so that the network predictions incorporate physical relationships as well as mathematical models to fit the data. The second method shows how partitioning the data into different regimes based on different physical processes, and training different networks for different regimes, improves the accuracy of predictions.

Combustion under stratified operating conditions in a direct-injection spark-ignition engine was investigated using simultaneous planar laser-induced fluorescence imaging of the fuel distribution (via 3-pentanone doped into the fuel) and the combustion products (via OH, which occurs naturally). The simultaneous images allow direct determination of the flame front location under highly stratified conditions where the flame, or product, location is not uniquely identified by the absence of fuel. The 3-pentanone images were quantified, and an edge detection algorithm was developed and applied to the OH data to identify the flame front position. The result was the compilation of local flame-front equivalence ratio probability density functions (PDFs) for engine operating conditions at 600 and 1200 rpm and engine loads varying from equivalence ratios of 0.89 to 0.32 with an unthrottled intake. Homogeneous conditions were used to verify the integrity of the method.

Transient operation of automobile engines is known to contribute significantly to regulated exhaust emissions, and is also an area of drivability concerns. Furthermore, many on-board diagnostic algorithms do not perform well during transient operation and are often temporarily disabled to avoid problems. The inability to quickly and repeatedly test engines during transient conditions in a laboratory setting limits researchers and development engineers ability to produce more effective and robust algorithms to lower vehicle emissions. To meet this need, members of the Powertrain Control Research Laboratory (PCRL) at the University of Wisconsin-Madison have developed a high-bandwidth, hydrostatic dynamometer system that will enable researchers to explore transient characteristics of engines and powertrains in the laboratory.

Automatic manufacturability analysis of injection moldings, sheet metal castings, stampings, forgings, etc., using knowledge-based heuristics depends on shape features, which are abstractions of the three dimensional (3D) geometric model of the parts. Conventional CAD systems do not explicitly contain shape feature information, therefore such information needs to be extracted from them. So far, extraction of shape features has been restricted to models with simple geometric shapes such as planar, cylindrical or conical shapes. Extending shape feature extraction to non-linear geometric models will allow Design For Manufacturability (DFM) analysis of non-linear models. This paper presents an approach to extract features from non-linear geometric models. The approach is based on abstract geometric entities called C-loops. The formation of a C-loop depends on a geometric entity called a silhouette. The C-loops are derived from the silhouette boundaries of an object.

The present work is aimed at developing a computational model of the interelectrode phenomena in thermionic energy converters which will be accurate over a very wide range of plasma conditions and operating modes. Previous models have achieved only moderate degrees of accuracy and, in a limited range, of validity. This limited range excludes a number of advanced thermionic devices, such as barium-cesium converters. The model under development promises improved accuracy in prediction of conventional devices and extension of predictive capability to advanced devices. The approach is to adapt the “Converted Scheme”, or CS method, to the cesium vapor plasma diode. This method, developed at the University of Wisconsin- Madison, is an extremely efficient algorithm for the solution of charged-particle kinetic equations and has been successfully used to simulate helium RF glow discharges.

At past ESV conferences, we have reported on a series of studies on how the driver's control performance is affected by vehicle steering response. These studies showed that a four-wheel steering system can reduce the delay in lateral acceleration response to steering action, which may result in better control performance of the driver. The present report examines the handling performance of an experimental vehicle fitted with a four-wheel steering system under a wider range of operating conditions. The studies were conducted using mathematical models and simulation of the driver-vehicle system, plus road tests. The findings indicate that the four-wheel steering system may provide better vehicle handling performance than a conventional two-wheel steering system. A vehicle incorporating this steering system may exhibit improved accident avoidance capabilities.