Search Results

Abstract Lane changes are stressful maneuvers for drivers, particularly during high-speed traffic flows. However, modeling driver’s lane-changing decision and implementation process is challenging due to the complexity and uncertainty of driving behaviors. To address this issue, this article presents a personalized Lane-Changing Model (LCM) for Advanced Driver Assistance System (ADAS) based on deep learning method. The LCM contains three major computational components. Firstly, with abundant inputs of Root Residual Network (Root-ResNet), LCM is able to exploit more local information from the front view video data. Secondly, the LCM has an ability of learning the global spatial-temporal information via Temporal Modeling Blocks (TMBs). Finally, a two-layer Long Short-Term Memory (LSTM) network is used to learn video contextual features combined with lane boundary based distance features in lane change events.

Abstract A kinematic modeling framework was established to predict status (position, displacement, velocity, acceleration, and shape) of a towing vehicle system with different driver inputs. This framework consists of three components: (1) a state space model to decide position and velocity for the vehicle system based on Newton’s second law; (2) an angular acceleration transferring model, which leads to a hypothesis that the each towed unit follows the same path as the towing vehicle; and (3) a polygon model to draw instantaneous polygons to envelop the entire system at any time point.

Abstract In this paper, an efficient lane detection using deep feature extraction method is proposed to achieve real-time lane detection in diverse road environment. The method contains three main stages: 1) pre-processing, 2) deep lane feature extraction and 3) lane fitting. In pre-processing stage, the inverse perspective mapping (IPM) is used to obtain a bird's eye view of the road image, and then an edge image is generated using the canny operator. In deep lane feature extraction stage, an advanced lane extraction method is proposed. Firstly, line segment detector (LSD) is applied to achieve the fast line segment detection in the IPM image. After that, a proposed adaptive lane clustering algorithm is employed to gather the adjacent line segments generated by the LSD method. Finally, a proposed local gray value maximum cascaded spatial correlation filter (GMSF) algorithm is used to extract the target lane lines among the multiple lines.

Abstract Potential collisions with oncoming traffic while turning left belong to the most safety-critical situations accounting for ~25% of all intersection crossing path crashes. A Left Turn Assist (LTA) was developed to reduce the number of crashes. Crucial for the effectiveness of the system is the design of the human-machine interface (HMI), i.e. defining how the system uses the calculated crash probability in the communication with the driver. A driving simulator study was conducted evaluating a warning strategy for two use cases: firstly, the driver comes to a stop before turning (STOP), and secondly, the driver moves on without stopping (MOVE). Forty drivers drove through three STOP and two MOVE scenarios. For the STOP scenarios, the study compared the effectiveness of an audio-visual warning with an additional brake intervention and a baseline. For the MOVE scenarios, the study analyzed the effectiveness of the audio-visual warning against a baseline.

Abstract In motorcycles, when they are traveling at medium to high speed, the roll stability is usually maintained by the restoration force generated by self-steering effect. However, when the vehicle is stationary or traveling in low speed, sufficient restoring force does not occur because some of the forces, such as centrifugal force, become small. In our study, we aimed at prototyping a motorcycle having a roll stability realized by a steering control when the vehicle is stationary or traveling in low speed. When we considered a mathematical control model to be applied, general models of four-degree-of-freedom had a critical inconvenience that the formulae include nonlinear second derivatives making them excessively complicated for deriving a practically applicable control method. Accordingly, we originally constructed a new control model which has equivalent two point masses (upper and lower from the vehicle’s center of gravity).

Abstract Since the tire inflation pressure has a significant influence on safety, comfort and environmental behavior of a vehicle, the choice of the optimal inflation pressure is always a conflict of aims. The development of a highly dynamic Tire Pressure Control System (TPCS) can reduce the conflict of minimal rolling resistance and maximal traction. To study the influence of the tire inflation pressure on longitudinal tire characteristics under laboratory conditions, an experimental sensitivity analysis is performed using a multivalent usable Corner Module Test Rig (CMTR) developed by the Automotive Engineering Group at Technische Universität Ilmenau. The test rig is designed to analyze suspension system and tire characteristics on a roller of the recently installed 4 chassis roller dynamometer. Camber angle, toe angle and wheel load can be adjusted continuously. In addition, it is possible to control the temperature of the test environment between −20 °C and +45 °C.

Abstract The advancement towards development of autonomy follows either the bottom-up approach of gradually improving and expanding existing Advanced Driver Assist Systems (ADAS) technology where the driver is present in the control loop or the top-down approach of directly developing autonomous vehicle hardware and software using alternative approaches without the driver present in the control loop. Most ADAS systems today fall under the classification of SAE Level 1 which is also referred to as the driver assistance level. The progression from SAE Level 1 to SAE Level 2 or partial automation involves the critical task of merging automated lateral control and automated longitudinal control such that the tasks of steering and acceleration/deceleration are not required to be handled by the driver under certain conditions [1].

Abstract Modern vehicles have incorporated numerous safety-focused advanced driver-assistance systems (ADAS) in the last decade including smart cruise control and object avoidance. In this article, we aim to go beyond using ADAS for safety and propose to use ADAS technology to enable predictive optimal energy management and improve vehicle fuel economy (FE). We combine ADAS sensor data with a previously developed prediction model, dynamic programming (DP) optimal energy management control, and a validated model of a 2010 Toyota Prius to explore FE. First, a unique ADAS detection scope is defined based on optimal vehicle control prediction aspects demonstrated to be relevant from the literature. Next, during real-world city and highway drive cycles in Denver, Colorado, a camera is used to record video footage of the vehicle environment and define ADAS detection ground truth. Then, various ADAS algorithms are combined, modified, and compared to the ground truth results.

Abstract The ability of senior citizens as well as other members of the general population to engage in an effective manner with technology is of increasing importance as new and innovative technologies become available. While recognizing the challenges that technologies can have on different populations, the ability to interact successfully with new technologies will, for seniors, have important consequences that can affect their quality of life and those of their families in numerous and important ways. This study, building upon previous research, examines the major dimensions of decision-making regarding attitudes toward autonomous vehicle technologies (ATVs) and their use. The study utilized data from a study of senior citizens in the Dallas-Fort Worth (DFW) area and compared the results with a sample of graduate students from a local university.

Abstract In Advanced Driver Assistant System (ADAS), the automotive radar is used to detect targets or obstacles around the vehicle. The procedure of Constant False Alarm Rate (CFAR) plays an important role in adaptive targets detection in noise or clutter environment. But in practical applications, the noise or clutter power is absolutely unknown and varies over the change of range, time and angle. The well-known cell averaging (CA) CFAR detector has a good detection performance in homogeneous environment but suffers from masking effect in multi-target environment. The ordered statistic (OS) CFAR is more robust in multi-target environment but needs a high computation power. Therefore, in this paper, a new two-dimension CFAR procedure based on a combination of Generalized Order Statistic (GOS) and CA CFAR named GOS-CA CFAR is proposed. Besides, the Linear Frequency Modulation Continuous Wave (LFMCW) radar simulation system is built to produce a series of rapid chirp signals.

Abstract Weather conditions such as rain, fog, snow, and dust can adversely impact sensing and perception, limit operational envelopes, and compromise the safety and reliability of advanced driver-assistance systems and autonomous vehicles. Physical testing of an autonomous system in a weather laboratory and on-road is costly and slow and exposes the system to only a limited set of weather conditions. To overcome the limitations of physical testing, a physics-based simulation workflow was developed by coupling computational fluid dynamics (CFD) with optical simulations of camera and lidar sensors. The computational data of various weather conditions can be rapidly generated by CFD and used to assess the impact of weather conditions on the sensors and perception algorithms.

Abstract This article describes the application of Reinforcement Learning (RL) with an embedded heuristic algorithm to a multi-objective hybrid vehicle optimization. A multi-objective optimization problem (MOP) is defined as a minimization of total energy consumption and trip time resulting from optimal control of vehicle speed over a known route. First, a computationally efficient heuristic optimization algorithm is formulated to solve the MOP for multiple traffic scenarios. Then, the off-line integration of RL is applied to the heuristic optimization algorithm process and utilized to solve the MOP. Finally, the online optimization capability of the machine learning algorithm is discussed, as well as its extension to the vehicle routing problem and the hybrid electric vehicle. The specific scenario investigated is where a generic vehicle begins a trip on a one-lane highway. The length of the highway and the number of vehicles and traffic signals on the road are generic as well.

Abstract Autonomous vehicles require precise knowledge of their position and orientation in all weather and traffic conditions for path planning, perception, control, and general safe operation. Here we derive these requirements for autonomous vehicles based on first principles. We begin with the safety integrity level, defining the allowable probability of failure per hour of operation based on desired improvements on road safety today. This draws comparisons with the localization integrity levels required in aviation and rail where similar numbers are derived at 10−8 probability of failure per hour of operation. We then define the geometry of the problem where the aim is to maintain knowledge that the vehicle is within its lane and to determine what road level it is on.

Abstract This work was supported jointly by the National Science Foundation of China under Grant No. 51875184 and the National key R&D programs, China New energy vehicles focus on special projects under Grant No. 2016YFB0100903-2.

Abstract Due to the transition of the driver to a passenger as well as the option of non-driving tasks, automated driving will necessitate adjustments of driving dynamics. In order to face higher comfort requirements and mitigate motion sickness not only horizontal dynamics but also vertical dynamics should be concerned. Therefore, we developed a novel control algorithm for active suspension systems, which takes the requirements of autonomous vehicles into account. Due to safety, cost reasons, and the unavailability of automated test vehicles, the controller was built up, tested, and tuned in simulation before final in-car testing. In this article we introduce a combined simulation and testing process for suspension control systems with focus on comfort measures. We successfully apply the method to the mentioned active suspension control algorithm with good accordance between simulation and measurement for low-frequency excitation.

Abstract Testing of advanced driver assistance system (ADAS) functions and automated driving (AD) technologies is an important step in their development, which cannot be addressed by the traditional automotive validation and verification processes. It has to be guaranteed before their deployment that the ADAS/AD functions exhibit appropriate and intended vehicle behavior under all possible scenarios. Particularly, analyzing the impact of ADAS/AD systems on traffic safety and efficiency is not trivial and is an active research topic. In the current article, a submicroscopic co-simulation framework is introduced consisting of microscopic traffic simulation, infrastructure communications, realistic vehicle dynamics, and corresponding ADAS functions.

Abstract Autonomous vehicle technology has become an unprecedented trend in the development of the automobile industry, which can ensure highly efficient use of resources, effectively improve the driving experience, and greatly reduces the driver’s burden. As one of the key technologies of autonomous vehicles, path planning has an important impact on the practical applications of autonomous vehicles. Planning a proper and efficient path is a prerequisite, which can improve the driving experience of autonomous vehicles. Therefore, in-depth research and development on applications of AI technology in path planning definitely have significant value in academic research. In this article, we will introduce a variety of path planning approaches for autonomous vehicles. We summarize the attributes of these path planning algorithms; simultaneously, we analyze the improvements to these algorithms. Then, we have a preliminary discussion on the applications in vehicle positioning and navigation.

Abstract Safety is always a crucial aspect of developing autonomous systems, and the motivation behind this project comes from the need to address the traffic crashes occurring globally on a daily basis. The present work studies the coexistence of the novel rule-based behavioral planning framework with the five key advanced driver assistance system (ADAS) features as proposed in this article to fulfill the safety requirements and enhance the comfort of the driver/passengers to achieve a receding-horizon autopilot. This architecture utilizes data from the sensor fusion and the prediction module for the prediction time horizon of 2 s iteratively, which is continuously moving forward (hence, the receding horizon), and helps the behavior planner understand the intent of other vehicles on the road in advance.

Abstract In this article, we propose a method of combining neural networks (NN) with nonlinear state-space models (SSM). Such model parts that are well understood can be integrated into the state space, while the NN can estimate such parts that are uncertain or hard to model. We apply the method to vehicle state estimation on a race track. Therefore, we derive a nonlinear two-track model with a scaled magic formula and adaptively estimate the tire parameters—stiffnesses and maximum friction potential—with the NN. The results show that the NN is able to reach an excellent estimation performance and generalizes over different model parameters, such as tire type, tread depth, surfaces conditions, and maneuvers. The trained model is furthermore integrated into an Extended Kalman Filter (EKF) to estimate the longitudinal speed, lateral speed, and yaw rate of the vehicle.

Abstract For safe and comfortable automated driving in the urban domain, especially in complex geometries as intersections, the prediction of surrounding traffic participants is fundamental. Several works in this field focus on predicting the behavior of vulnerable road users (VRU) at crossings. However, no approaches were found dealing with predicting the interaction between turning vehicles giving right of way or cooperating with VRU, which is substantial for the trajectory planning of following vehicles. Infrastructural sensor data from an intersection in Germany enables the development of a prediction concept for vehicles interacting with VRU. Our studies show that the original criteria for classifying an interaction between vehicles and VRU—the post-encroachment time (PET)—is not suitable as ground truth criteria for the aimed prediction. Instead, a clustering-based labelling approach with k-means shows promising results in trajectory pattern distinction.