Search Results

A real-time control is proposed for Plug-in Hybrid Electric Vehicles (PHEVs) based on the optimal Dynamic Programming (DP) trajectories in this study. Firstly, the DP is used to solve the Driving Cycle to obtain the optimal trajectories and controls, and the Model-Based Calibration tool (MBC) is used to generate the optimal Maps for the given optimal trajectories. Secondly, a Feedback Energy Management System (FMES) is built with State of Charge (SoC) as the feedback variable, which takes into account the Charge and Discharge Reaction (CDR) of the battery.

With the advent of autonomous vehicles, the human driver’s attention will slowly be relinquished from the driving task. It will allow drivers to participate in more non-driving related activities, such as engaging with information and entertainment systems. However, the automated driving system would need to notify the driver of upcoming points-of-interest on the road when the driver’s attention is focused on their screen rather than on the road or driving display. In this paper, we investigated whether providing directional alerts for an upcoming point-of-interest (POI) in or around the user’s active screen can augment their ability in relocating their visual attention to the POI on the road when traveling in a vehicle with Conditional Driving Automation. A user study (N = 15) was conducted to compare solutions for alerts that presented themselves in the participants’ central and peripheral field of view.

Currently, most of vehicle recognition methods are realized by deep convolutional neural networks (DCNNs) with input of images directly as training data. Due to the factor of perspective distortion and scale change of images taken by monocular camera, a large number of multi-scale images need to be used for training, and physical information of vehicles cannot be obtained at the same time. In order to improve the above problems, we present a method of vehicle recognition based on unpacking 3D bounding boxes in this paper. Firstly, camera calibration information and geometric constraints are used to build 3D bounding boxes around vehicles in monocular projection. Then, the 3D bounding boxes are unpacked to obtain 3D normalized spatial data without perspective distortion. Finally, VGG-16 is chosen as the backbone of our network, the output of which can be divided into five common vehicle types including hatchback, sedan, SUV, truck and bus.