Search Results

Traditional autonomous vehicle perception subsystems that use onboard sensors have the drawbacks of high computational load and data duplication. Infrastructure-based sensors, which can provide high quality information without the computational burden and data duplication, are an alternative to traditional autonomous vehicle perception subsystems. However, these technologies are still in the early stages of development and have not been extensively evaluated for lane detection system performance. Therefore, there is a lack of quantitative data on their performance relative to traditional perception methods, especially during hazardous scenarios, such as lane line occlusion, sensor failure, and environmental obstructions.

Lane detection plays a critical role in autonomous vehicles for safe and reliable navigation. Lane detection is traditionally accomplished using a camera sensor and computer vision processing. The downside of this traditional technique is that it can be computationally intensive when high quality images at a fast frame rate are used and has reliability issues from occlusion such as, glare, shadows, active road construction, and more. This study addresses these issues by exploring alternative methods for lane detection in specific scenarios caused from road construction-induced lane shift and sun glare. Specifically, a U-Net, a convolutional network used for image segmentation, camera-based lane detection method is compared with a radar-based approach using a new type of sensor previously unused in the autonomous vehicle space: radar retro-reflectors.

Evaluating real-world hazards associated with perception subsystems is critical in enhancing the performance of autonomous vehicles. The reliability of autonomous vehicles perception subsystems are paramount for safe and efficient operation. While current studies employ different metrics to evaluate perception subsystem failures in autonomous vehicles, there still exists a gap in the development and emphasis on engineering requirements. To address this gap, this study proposes the establishment of engineering requirements that specifically target real-world hazards and resilience factors important to AV operation, using High-Definition Maps, Global Navigation Satellite System, and weather sensors. The findings include the need for engineering requirements to establish clear criteria for a high-definition maps functionality in the presence of erroneous perception subsystem inputs which enhances the overall safety and reliability of the autonomous vehicles.

Accurate perception of the driving environment and a highly accurate position of the vehicle are paramount to safe Autonomous Vehicle (AV) operation. AVs gather data about the environment using various sensors. For a robust perception and localization system, incoming data from multiple sensors is usually fused together using advanced computational algorithms, which historically requires a high-compute load. To reduce AV compute load and its negative effects on vehicle energy efficiency, we propose a new infrastructure information source (IIS) to provide environmental data to the AV. The new energy–efficient IIS, chip–enabled raised pavement markers are mounted along road lane lines and are able to communicate a unique identifier and their global navigation satellite system position to the AV. This new IIS is incorporated into an energy efficient sensor fusion strategy that combines its information with that from traditional sensor.

Contemporary ADS and ADAS localization technology utilizes real-time perception sensors such as visible light cameras, radar sensors, and lidar sensors, greatly improving transportation safety in sufficiently clear environmental conditions. However, when lane lines are completely occluded, the reliability of on-board automated perception systems breaks down, and vehicle control must be returned to the human driver. This limits the operational design domain of automated vehicles significantly, as occlusion can be caused by shadows, leaves, or snow, which all occur in many regions. High-definition map data, which contains a high level of detail about road features, is an alternative source of the required lane line information. This study details a novel method where high-definition map data are processed to locate fully occluded lane lines, allowing for automated path planning in scenarios where it would otherwise be impossible.

Modern vehicles use automated driving assistance systems (ADAS) products to automate certain aspects of driving, which improves operational safety. In the U.S. in 2020, 38,824 fatalities occurred due to automotive accidents, and typically about 25% of these are associated with inclement weather. ADAS features have been shown to reduce potential collisions by up to 21%, thus reducing overall accidents. But ADAS typically utilize camera sensors that rely on lane visibility and the absence of obstructions in order to function, rendering them ineffective in inclement weather. To address this research gap, we propose a new technique to estimate snow coverage so that existing and new ADAS features can be used during inclement weather. In this study, we use a single camera sensor and historical weather data to estimate snow coverage on the road. Camera data was collected over 6 miles of arterial roadways in Kalamazoo, MI.

Practical applications of recently developed sensor fusion algorithms perform poorly in the real world due to a lack of proper evaluation during development. Existing evaluation metrics do not properly address a wide variety of testing scenarios. This issue can be addressed using proactive performance measurements such as the tools of resilience engineering theory rather than reactive performance measurements such as root mean square error. Resilience engineering is an established discipline for evaluating proactive performance on complex socio-technical systems which has been underutilized for automated vehicle development and evaluation. In this study, we use resilience engineering metrics to assess the performance of a sensor fusion algorithm for vehicle localization. A Kalman Filter is used to fuse GPS, IMU and LiDAR data for vehicle localization in the CARLA simulator.

Perception in adverse weather conditions is one of the most prominent challenges for automated driving features. The sensors used for mid-to-long range perception most impacted by weather (i.e., camera and LiDAR) are susceptible to data degradation, causing potential system failures. This research series aims to better understand sensor data degradation characteristics in real-world, dynamic environmental conditions, focusing on adverse weather. To achieve this, a dataset containing LiDAR (Velodyne VLP-16) and camera (Mako G-507) data was gathered under static scenarios using a single vehicle target to quantify the sensor detection performance. The relative position between the sensors and the target vehicle varied longitudinally and laterally. The longitudinal position was varied from 10m to 175m at 25m increments and the lateral position was adjusted by moving the sensor set angle between 0 degrees (left position), 4.5 degrees (center position), and 9 degrees (right position).

Robust sensor fusion is a key technology for enabling the safe operation of automated vehicles. Sensor fusion typically utilizes inputs of cameras, radars, lidar, inertial measurement unit, and global navigation satellite systems, process them, and then output object detection or positioning data. This paper will focus on sensor fusion between the camera, radar, and vehicle wheel speed sensors which is a critical need for near-term realization of sensor fusion benefits. The camera is an off-the-shelf computer vision product from MobilEye and the radar is a Delphi/Aptive electronically scanning radar (ESR) both of which are connected to a drive-by-wire capable vehicle platform. We utilize the MobilEye and wheel speed sensors to create a dynamic region of interest (DROI) of the drivable region that changes as the vehicle moves through the environment.

Today’s Advanced Driver Assistance Systems (ADAS) predominantly utilize cameras to increase driver and passenger safety. Computer vision, as the enabler of this technology, extracts two key environmental features: the drivable region and surrounding objects (e.g., vehicles, pedestrians, bicycles). Lane lines are the most common characteristic extracted for drivable region detection, which is the core perception task enabling ADAS features such as lane departure warnings, lane-keeping assistance, and lane-centering. However, when subject to adverse weather conditions (e.g., occluded lane lines) the lane line detection algorithms are no longer operational. This prevents the ADAS feature from providing the benefit of increased safety to the driver. The performance of one of the leading computer vision system providers was tested in conditions of variable snow coverage and lane line occlusion during the 2020-2021 winter in Kalamazoo, Michigan.

This study presents experimental and numerical examination of directly injected (DI) propane and iso-octane, surrogates for liquified petroleum gas (LPG) and gasoline, respectively, at various engine like conditions with the overall objective to establish the baseline with regards to fuel delivery required for future high efficiency DI-LPG fueled heavy-duty engines. Sprays for both iso-octane and propane were characterized and the results from the optical diagnostic techniques including high-speed Schlieren and planar Mie scattering imaging were applied to differentiate the liquid-phase regions and the bulk spray phenomenon from single plume behaviors. The experimental results, coupled with high-fidelity internal nozzle-flow simulations were then used to define best practices in CFD Lagrangian spray models.

Operation of an autonomous vehicle (AV) carries risk if it acts on inaccurate information about itself or the environment. The perception system is responsible for interpreting the world and providing the results to the path planning and other decision systems. The perception system performance is a result of the operating state of the sensors, e.g. is a sensor in fault or being adversely affected by the weather or environmental conditions, and approach to sensor measurement interpretation. We propose a trailing horizon switched system observer that minimizes the difference between reference tracking values developed from sensor fusion performed at an upper level and the values from a potentially faulty sensor based upon the convex combination of different sensor observation model outputs; the sensor observations models are associated with different sensor operating errors.

Advanced compression ignition (ACI) operation in gasoline direct injection (GDI) engines is a promising concept to reduce fuel consumption and emissions at part load conditions. However, combustion phasing control and the limited operating range in ACI mode are a perennial challenge. In this study the combined impact of fuel properties and engine control strategies in ACI operation are investigated. A design of experiments method was implemented using a three level orthogonal array to determine the sensitivity of engine control parameters on the engine load, combustion noise and stability under low load ACI operation for three RON 98 gasoline fuels, each exhibiting disparate chemical composition. Furthermore, the thermodynamic state of the compression histories was studied with the aid of the pressure-temperature framework.

Autonomous vehicle technology has the potential to improve the safety, efficiency, and cost of our current transportation system by removing human error. With sensors available today, it is possible for the development of these vehicles, however, there are still issues with autonomous vehicle operations in adverse weather conditions (e.g. snow-covered roads, heavy rain, fog, etc.) due to the degradation of sensor data quality and insufficiently robust software algorithms. Since autonomous vehicles rely entirely on sensor data to perceive their surrounding environment, this becomes a significant issue in the performance of the autonomous system. The purpose of this study is to collect sensor data under various weather conditions to understand the effects of weather on sensor data. The sensors used in this study were one camera and one LiDAR. These sensors were connected to an NVIDIA Drive Px2 which operated in a 2019 Kia Niro.

Currently there is a significant research effort being made in gasoline spark/ignition (SI) engines to understand and reduce cycle-to-cycle variations. One of the phenomena that presents this cycle-to-cycle variation is combustion knock, which also happens to have a very stochastic behavior in modern SI engines. Conversely, the CFR octane rating engine presents much more repeatable combustion knock activity. The aim of this study is to assess the impact of fuel composition on the cycle to cycle variation of the pressure and timing of end gas autoignition. The variation of cylinder conditions at the timing of end-gas autoignition (knock point) for a wide selection of cycle ensembles have been analyzed for several constant RON 98 fuels on the CFR engine, as well as in a modern single-cylinder gasoline direct injection (GDI) SI engine operated at RON-like intake conditions.

This study is a continuation of previous work assessing the robustness of a Cummins XPI common rail injection system operating with gasoline-like fuel. All the hardware from the original study was retained except for the high pressure pump head and check valves which were replaced due to cavitation damage. An additional 400 hour NATO cycle was run on the refurbished fuel system to achieve a total exposure time of 800 hours and detect any other significant failure modes. As in the initial investigation, fuel system parameters including pressures, temperatures and flow rates were logged on a test bench to monitor performance over time. Fuel and lubricant samples were taken every 50 hours to assess fuel consistency, metallic wear, and interaction between fuel and oil. High fidelity driving torque and flow measurements were made to compare overall system performance when operating with both diesel and light distillate fuel.

There are many challenges in implementing grid aware plug-in electric vehicle (PEV) charging systems with local load control. New opportunities for innovative load control were created as a result of changes to the 2014 National Electric Code (NEC) about automatic load control definitions for EV charging infrastructure. Stakeholders in optimized dispatch of EV charging assets include the end users (EV drivers), site owner/operators, facility managers and utilities. NEC definition changes allow for ‘over subscription’ of more potential EV charging station load than can be continuously supported if the total load at any time is within the supply system safety limit. Local load control can be implemented via compact submeter(s) with locally hosted control algorithms using direct communication to the managed electric vehicle supply equipment (EVSE).

This work presents a methodology to determine the off-cycle fuel economy benefit of a 2-Layer HVAC system which reduces ventilation and heat rejection losses of the heater core versus a vehicle using a standard system. Experimental dynamometer tests using EPA drive cycles over a broad range of ambient temperatures were conducted on a highly instrumented 2016 Lexus RX350 (3.5L, 8 speed automatic). These tests were conducted to measure differences in engine efficiency caused by changes in engine warmup due to the 2-Layer HVAC technology in use versus the technology being disabled (disabled equals fresh air-considered as the standard technology baseline). These experimental datasets were used to develop simplified response surface and lumped capacitance vehicle thermal models predictive of vehicle efficiency as a function of thermal state.

The fuel spray behavior in the near nozzle region of a gasoline injector is challenging to predict due to existing pressure gradients and turbulences of the internal flow and in-nozzle cavitation. Therefore, statistical parameters for spray characterization through experiments must be considered. The characterization of spray velocity fields in the near-nozzle region is of particular importance as the velocity information is crucial in understanding the hydrodynamic processes which take place further downstream during fuel atomization and mixture formation. This knowledge is needed in order to optimize injector nozzles for future requirements. In this study, the results of three experimental approaches for determination of spray velocity in the near-nozzle region are presented. Two different injector nozzle types were measured through high-speed shadowgraph imaging, Laser Doppler Anemometry (LDA) and X-ray imaging.

Modeling plume interaction and collapse for direct-injection gasoline sprays is important because of its impact on fuel-air mixing and engine performance. Nevertheless, the aerodynamic interaction between plumes and the complicated two-phase coupling of the evaporating spray has shown to be notoriously difficult to predict. With the availability of high-speed (100 kHz) Particle Image Velocimetry (PIV) experimental data, we compare velocity field predictions between plumes to observe the full temporal evolution leading up to plume merging and complete spray collapse. The target “Spray G” operating conditions of the Engine Combustion Network (ECN) is the focus of the work, including parametric variations in ambient gas temperature. We apply both LES and RANS spray models in different CFD platforms, outlining features of the spray that are most critical to model in order to predict the correct aerodynamics and fuel-air mixing.