Search Results

The automotive industry is heading towards the path of autonomy with the development of autonomous vehicles. An autonomous vehicle consists of two main components. The first is the software which is responsible for the decision-making capabilities of the system. The second is the hardware which encompasses all aspects of the physical vehicle which are responsible for vehicle motion such as the engine, brakes and steering subsystems along with their corresponding controls. This component forms the basis of the autonomous vehicle platform. For SAE Level 4 autonomous vehicles, where an automated driving system is responsible for all the dynamics driving tasks including the fallback driving performance in case of system faults, redundant mechanical systems and controls are required as part of the autonomous vehicle platform since the driver is completely out of the loop with respect to driving.

With the recent advancements in sensing and processing capabilities of consumer mobile devices (e.g., smartphone, tablet, etc.), they are becoming attractive choices for pervasive computing applications. Always-on monitoring of human movement patterns is one of those applications that has gained a lot of importance in the field of mobility and transportation research. Automatic detection of the current transportation mode (e.g., walking, biking, riding a shuttle, etc.) of a consumer using data from their smartphone sensors enables delivering of a number of customized services for multi-modal journey planning. Most accurate models for automatic mode detection are trained with supervised learning algorithms. In order to achieve high accuracy, the training datasets need to be sufficiently large, diverse, and correctly labeled.

Road Load Data is an important source of information for quality improvements. Vehicle and component load information, such as driver behaviour and other characteristics of vehicle use in real world conditions, are the basis for many engineering tasks, including fuel economy and life-time optimization. This new approach in road load data acquisition is based on the increasing existence of data bus networks in modern vehicles, as well as further improvements in data acquisition technologies. The applications of which alllow smart, inobtrusive solutions and, increasingly, the collection of real world customer data. This methodology of data collection leads to a significant alteration of the company-wide data base, relevant for future engineering efforts. The growing share of real world customer data will result in a more optimized and customer-orientated range of solutions, for all areas of vehicle engineering.

One of the ways to reduce city congestion is to balance the traffic flow on the road network and maximally utilize all road capacities. There are examples showing that, if the drivers are not competitive but cooperative, the road network usage efficiency and the traffic conditions can be improved. This motivates the idea of designing a cooperative routing algorithm to benefit most vehicles on the road. This paper presents a semi-cooperative social routing algorithm for large transportation network with predictive traffic density information. The goal is to integrate a cooperative scheme into the individual routing and achieve short traveling time not only for the traveler itself, but also for all vehicles in the road network. The most important concept of this algorithm is that the route is generated with the awareness of the total travel time added to all other vehicles on the road due to the increased congestion.

As intelligent automated vehicle technologies evolve, there is a greater need to understand and define the role of the human user, whether completely hands-off (L5) or partly hands-on. At all levels of automation, the human occupant may feel anxious or ill-at-ease. This may reflect as higher stress/workload. The study in this paper further refines how perceived workload may be determined based on occupant physiological measures. Because of great variation in individual personalities, age, driving experiences, gender, etc., a generic model applicable to all could not be developed. Rather, individual workload models that used physiological and vehicle measures were developed.

Recommender systems guide a user to useful objects in a large space of possible options in a personalized way. In this paper, we study recommender systems for vehicles. Compared to previous research on recommender systems in other domains (e.g., movies or music), there are two major challenges associated with recommending vehicles. First, typical customers purchase fewer cars than movies or pieces of music. Thus, it is difficult to obtain rich information about a customer’s vehicle purchase history. Second, content information obtained about a customer (e.g., demographics, vehicle preferences, etc.) is also difficult to acquire during a relatively short stay in a dealership. To address these two challenges, we propose an interactive vehicle recommender system based a novel machine learning method that integrates decision trees and multi-armed bandits. Decision tree learning effectively selects important questions to ask the customer and encodes the customer's key preferences.

Adaptive Cruise Control (ACC) is becoming a common feature in modern day vehicles with the advancement of Advanced Driver Assist Systems (ADAS). Simultaneously, Hardware-in-the-Loop (HIL) simulation has emerged as a major component of the automotive product development cycle as it can accelerate product development and validation by supplementing in-vehicle testing. Specifically, HIL simulation has become an integral part of the controls development and validation V-cycles by enabling rapid prototyping of control software for Electronic Control Units (ECUs). Traditionally, ACC algorithms have been validated on a system or subsystem HIL bench with the ACC ECU in the loop such that the HIL bench acts as the host or trailing vehicle with the target or preceding vehicle usually simulated using as an object that follows a pre-defined motion profile.

There has been a growing concern about increasing vehicle-mile traveled (VMT) associated with deadhead trips for dynamic rideshare services, particularly with the emergence of Shared Autonomous Vehicle (SAV) services. Studies in the literature on repositioning strategies have been limited to synthetic or small-scale study areas. This study considers a large-scale computational experiment involving a New York City study area with a network of 16,782 nodes and 23,337 links with 662,455 potential travelers from the 2016 Yellow Taxi data. We investigate the potential to reduce VMT and deadhead miles for dynamic rideshare operations combined with vehicle repositioning strategies. Three repositioning strategies are evaluated: (1) Roaming around areas with higher pickup probabilities to maximize the chance of picking up passengers, (2) Staying at curb side after completing trips, and (3) Repositioning to depots to minimize deadhead trips.