Search Results

Autonomous farming has gained a vast interest due to the need for increased farming efficiency and productivity as well as reducing operating cost. Technological advancement enabled the development of Autonomous Driving (AD) features in unstructured environments such as farms. This paper discusses an approach of utilizing satellite images to estimate the drivable areas of agriculture fields with the aid of LiDAR sensor data to provide the necessary information for the vehicle to navigate autonomously. The images are used to detect the field boundaries while the LiDAR sensor detects the obstacles that the vehicle encounters during the autonomous driving as well as its type. These detections are fused with the information from the satellite images to help the path planning and control algorithms in making safe maneuvers. The image and point cloud processing algorithms were developed in MATLAB®/C++ software and implemented within the Robot Operating System (ROS) middleware.

For car-following models, the car-following characteristics differ depending on the vehicle type, such as passenger cars, motorcycles, and trucks. Therefore, constructing a model for each category is essential. To that end, various modeling methods have been proposed; however, herein, we particularly focused on the long short-term memory (LSTM), which is the best method for forecasting long-term time-series data.[1, 2] The objective of this study was to construct a car-following model for each vehicle category using the LSTM and to evaluate the model accuracy for each vehicle category. In this study, US-101 and I-80 data provided by the next-generation simulation (NGSIM), which is based on natural traffic flow data, were used. In the NGSIM, only car-following situations were selected as car-following data, and these were classified into the vehicle type: motorcycles, passenger cars, and trucks.

Technology is ever advancing in the world around us, and it is no different when it comes to data acquisition systems used in accident reconstruction. In 2016, the SAE publication “Data Acquisition Using Smart Phone Applications,” Neale et al. evaluated the accuracy of basic fitness applications in tracking position within the smart phone itself [1]. In 2018, a follow up publication “Mid-Range Data Acquisition Units Using GPS and Accelerometers” tested the Harry’s Lap TimerTM application for use in smart phones and compared the data to the Race Logic VBOX [2]. In this paper, another data acquisition system, the MoTeC C185, was tested. The MoTeC C185 data logger contains an internal 3-axis accelerometer and was also equipped with an external Syvecs 50Hz GPS Module with 6-axis accelerometer. A test vehicle was instrumented with the MoTeC C185, Race Logic VBOX, and Harry’s Lap TimerTM.

This article presents an autonomous steering control scheme for articulated heavy vehicles (AHVs). Despite economic and environmental benefits in freight transportation, lateral stability is always a concern for AHVs in high-speed highway operations due to their multi-unit vehicle structures, and high centers of gravity (CGs). In addition, North American harsh winter weather makes the lateral stability even more challenging. AHVs often experience amplified lateral motions of trailing vehicle units in high-speed evasive maneuvers. AHVs represent a 7.5 times higher risk than passenger cars in highway operation. Human driver errors cause about 94% of traffic collisions. However, little attention has been paid to autonomous steering control of AHVs.

Autonomous truck with modular chassis has the characteristics of high driving flexibility and strong load capacity. It can be equipped with different numbers of modular chassis according to the task requirements. The application of autonomous truck can solve the problems of traffic accidents and shortage of drivers effectively, which is the development trend of trucks in the future. For the collision-free trajectory planning problem of dual-modular chassis autonomous truck, this paper designs a hierarchical local trajectory planner that combines the artificial potential field method with polynomial curve fitting method. This planner plans the center of mass trajectory firstly, and then generates the modular chassis trajectories according to the position relationship between the center of mass and the chassis.

Unmanned Ground Vehicle (UGV) has a wide range of applications in the military, agriculture, firefighting and other fields. Path planning, as a key aspect of autonomous driving technology, plays an essential role for UGV to accomplish the established driving tasks. At present, there are many global path planning algorithms in grid maps on unstructured roads, while general grid maps do not consider the specific elevation or ground type difference of each grid, and unstructured roads are generally considered as flat and open roads. On the contrary, the unmanned off-road is always a bumpy road with undulating terrain, and meanwhile, the landform is complex and the types of features are diverse. In order to ensure the safety and improve the efficiency of autonomous driving of UGV in off-road environment, this paper proposes a global off-road path planning method for UGV based on the raw image of remote sensing map. Firstly, the raw image is gridded.

Image segmentation has historically been a technique for analyzing terrain for military autonomous vehicles. One of the weaknesses of image segmentation from camera data is that it lacks depth information, and it can be affected by environment lighting. Light detection and ranging (LiDAR) is an emerging technology in image segmentation that is able to estimate distances to the objects it detects. One advantage of LiDAR is the ability to gather accurate distances regardless of day, night, shadows, or glare. This study examines LiDAR and camera image segmentation fusion to improve an advanced driver-assistance systems (ADAS) algorithm for off-road autonomous military vehicles. The volume of points generated by LiDAR provides the vehicle with distance and spatial data surrounding the vehicle.

Autonomous vehicles (AV) are sophisticated systems comprising various sensors, powerful processors, and complex data processing algorithms that navigate autonomously to their respective goals. Out of several functions performed by an AV, one of the most important is developing situational intelligence to predict collision-free future trajectories. As an AV operates in environments consisting of various entities, such as other AVs, human-driven vehicles, and static obstacles, developing situational intelligence will require a collaborative approach. The recent developments in artificial intelligence (AI) and deep learning (DL) relating to AVs have shown that DL-based models can take advantage of information sharing and collaboration to develop such intelligence.

The new generation vehicles these days are managed by networked controllers. A large portion of the networks is planned with more security which has recently roused researchers to exhibit various attacks against the system. This paper talks about the liabilities of the Controller Area Network (CAN) inside In-vehicle communication protocol and a few potentials that could take due advantage of it. Moreover, this paper presents a few security measures proposed in the present examination status to defeat the attacks. In any case, the fundamental objective of this paper is to feature a comprehensive methodology known as Intrusion Detection System (IDS), which has been a significant device in getting network data in systems over many years. To the best of our insight, there is no recorded writing on a through outline of IDS execution explicitly in the CAN transport network system.

Classic vehicle production had limitations in bringing the driving commands to the actuators for vehicle motion (engine, steering and braking). Steering columns, hydraulic tubes or steel cables needed to be placed between the driver and actuator. Change began with the introduction of e-gas systems. Mechanical cables were replaced by thin, electric signal wires. The technical solutions and legal standardizations for addressing the steering and braking systems, were not defined at this time. Today, OEMs are starting E/E-Architecture transformations for manifold reasons and now have the chance to remove the long hydraulic tubes for braking and the solid metal columns used for steering. X-by-wire is the way forward and allows for higher Autonomous Driving (AD) levels for automated driving vehicles. This offers new opportunities to design the vehicle in-cabin space. This paper will start with the introduction of x-by-wire technologies.

The purpose of this study was to construct driver models using long short-term memory (LSTM) in car-following situations, where other vehicles change lanes and cut in front of the ego vehicle (EGV). The development of autonomous vehicle systems (AVSs) using personalized driver models based on the individual driving characteristics of drivers is expected to reduce their discomfort with vehicle control systems. The driving characteristics of human drivers must be considered in such AVSs. In this study, we experimentally measured data from the EGV and other vehicles using a driving simulator consisting of a six-axis motion device and turntable. The experimental scenario simulated a traffic congestion scenario on a straight section of a highway, where a cut-in vehicle (CIV) changed lanes from an adjacent lane and entered in between the EGV and preceding vehicle (PRV).

Platoon is a system that connects vehicles through vehicle-to-vehicle (V2V) communication technology to maintain a short distance between vehicles while driving on the road. To improve fuel efficiency, many automotive original equipment manufacturers (OEMs) are interested in developing and demonstrating real-world platoon system. However, it is hard for heavy duty trucks to develop this system due to the difficulty of maintaining the targeted intervehicle distance not only for fuel efficiency but also for safety in case of emergency braking. Because of this critical safety issue in the emergency situation, the platoon system for heavy duty trucks can be hardly demonstrated or tested in real vehicle environment. The relatively complex system and the slow response characteristic of commercial vehicles makes this even more difficult.

This paper presents the results of five additional instrumented snowmobile crash tests and the updated findings when combined with the dataset from our previous publication (SAE# 2021-01-0876). This additional work expands on, and aims to better the understanding of, snowmobile collision dynamics and the severity of real-world collisions, given that minimal crash test data exists for snowmobiles compared to passenger vehicles. In the five new tests, the test snowmobile was accelerated forward into a pole, made from a tree trunk 38 cm in diameter. The first two pole impacts were completed at low-speed (i.e., less than 15 km/h) to determine a damage threshold. The subsequent two tests targeted 40 km/h impact speeds, and the last test targeted an impact speed of 80 km/h.

Prior research has tested the validity of Cycles Engine render in Blender for the creation of physically correct lighting models; however, a research gap still exists in examining the use of Arnold render engine in 3DS Max for accident reconstruction and other forensic settings [1]. Specifically, the process presented in this paper utilizes the Arnold render engine within 3DS-Max to analyze the lighting models. Arnold is a physically-based render (PBR) engine and can be used to recreate an accident scene geometry and lighting conditions. The goal is to create light sources within Arnold that represent the real-world light sources. The light sources in Arnold are quantified by several variables, including intensity, color, and size. The intensity and size variables determine the self-emitted radiance of the light source and require further explanation to determine the relationship between these variables in Arnold and real-world lighting quantities.

Autonomous truck and trailer configurations face challenges when operating in reverse due to the lack of sensing on the trailer. It is anticipated that sensor packages will be installed on existing trailers to extend autonomous operations while operating in reverse in uncontrolled environments, like a customer's loading dock. Power Line Communication (PLC) between the trailer and the tractor cannot support high bandwidth and low latency communication. This paper explores the impact of using Ethernet or a wireless medium for commercial trailer-tractor communication on the lifecycle and operation of trailer electronic control units (ECUs) from a Systems Engineering perspective to address system requirements, integration, and security. Additionally, content-based and host-based networking approaches for in-vehicle communication, such as Named Data Networking (NDN) and IP-based networking are compared.

Heavy-duty diesel trucking is responsible for 25%-30% of the road transportation CO2 emissions in North America. Retrofitting class-8 trucks with a complementary hydrogen fuelling system makes it possible to co-combust hydrogen and diesel in the existing internal combustion engine (ICE), thus minimizing the costs associated with switching to non-ICE platforms and reducing the barrier for the implementation of low-carbon gaseous fuels such as hydrogen. This retrofitting approach is evaluated based on the exhaust emissions of a converted truck with several thousand kilometres of road data. The heavy-duty truck used here was retrofitted with an air-intake hydrogen injection system, onboard hydrogen storage tanks, and a proprietary hydrogen controller enabling it to operate in hydrogen-diesel co-combustion (HDC) mode.

Published data relevant to snowmobile crash reconstruction is comparatively limited, especially pertaining to mountain snowmobiling and riding in deep snow. Snowmobiling is a unique motorsport activity as it requires substantial rider input and physical interaction to properly control the vehicle. The added complexities of varying slope angle and snow depth in mountain terrain make application of test data from testing done on flat surfaces less useful when applied to sloped terrain analysis. New data from testing performed in deep snow conditions on various slopes is presented in this paper. Acceleration tests were performed using two late model mountain snowmobiles from a stop on various slope angles. Additional related factors such as snow density, trenching, and snow mass momentum exchange are also discussed. Comparison of these test results to previously published snowmobile testing data advances the understanding of snowmobile acceleration parameters into mountain terrain.

Liquefied Petroleum Gas (LPG), as a common alternative fuel for internal combustion engines is currently widespread in use for fleet vehicles. However, a current majority of the LPG-fueled engines, uses port-fuel injection that offers lower power density when compared to a gasoline engine of equivalent displacement volume. This is due to the lower molecular weight and higher volatility of LPG components that displaces more air in the intake charge due to the larger volume occupied by the gaseous fuel. LPG direct-injection during the closed-valve portion of the cycle can avoid displacement of intake air and can thereby help achieve comparable gasoline-engine power densities. However, under certain engine operating conditions, direct-injection sprays can collapse and lead to sub-optimal fuel-air mixing, wall-wetting, incomplete combustion, and increased pollutant emissions.

Proton exchange membrane fuel cells are promising in the application of new energy vehicles and other fields. The performance test and analysis are critical components of the fuel cell research. A general procedure for data processing model and performance analysis method of fuel cell stacks were introduced based on a demonstration project of commercial fuel cell buses in Shanghai. To build the data processing model, the fixed node method was used to extract the effective operating data of fuel cell stack from the vehicle database. After data cleaning and conversion, taking fuel cell reversible recession into account, the processed data can be obtained. Data that had been processed and a semi-empirical model were used to accurately fit the polarization curve. Polarization curve is a pivotal approach to describe the performance of fuel cells.

The paper presents an approach used to generate a customer-oriented drive cycle using the MATLAB-based drive cycle generation tool for EVs developed by Isuzu Technical Center of America. The drive cycle generation tool extracts important features from the customer vehicle data and compares it with the globally used pre-existing candidate cycle to generate a representative drive cycle. The tool can read multiple file formats of preprocessed data or raw data from the vehicle telematics systems. This data is then processed using a unique and efficient algorithm developed by the Isuzu engineers, calculating seven important vehicle dynamic parameters. These selected parameters are compared with the pre-existing candidate cycles used across the globe in multiple iterations to generate a custom representative drive cycle that best fits real-world customer driving behavior.