Search Results

Automated driving is currently one of the most active areas of research worldwide. While the general progress in developing specific algorithms for perception, planning and control tasks is very advanced, testing and validation of the resulting functions is still challenging due to the large number of possible scenarios and generation of ground-truth. Currently, real world testing and simulations are used in combination to overcome some of these challenges. While real world testing does not suffer from imperfect sensor models and environments, it is expensive, slow and not accurately repeatable and therefore unable to capture all possible scenarios. However, simulation models are not sophisticated enough to fully replace real world testing. In this paper, we propose a workflow that is capable of augmenting real sensor-level data with simulated sensor data.

For Automated Vehicles (AVs) to be successful, they must integrate into society in a way that makes everyone confident in how AVs work to serve people and their communities. This integration requires that AVs communicate effectively, not only with other vehicles, but with all road users, including pedestrians and cyclists. One proposed method of AV communication is through an external human-machine interface (eHMI). While many studies have evaluated eHMI solutions, few have considered their compliance with relevant Federal Motor Vehicle Safety Standards (FMVSS) and their scalability. This study evaluated the effectiveness of a lightbar eHMI to communicate AV intent by measuring user comprehension of the eHMI and its impact on pedestrians’ trust and acceptance of AVs.

The interaction of Connect and Automated vehicles (CAV) with regular vehicles in the traffic stream has been extensively researched. Most studies, however, focus on calibrating driver behavior models for CAVs based on various levels of automation and driver aggressiveness. Other related studies largely focus on the coordination of CAVs and infrastructure like traffic signals to optimize traffic. However, the effects of different strategic flow management of CAVs in the traffic stream in the comparative scenario-based analysis is understudied. Thus, this study develops a framework and simulations for integrating CAVs in a corridor section. We developed a calibrated model with CAVs for a corridor section in Corktown, Michigan, and simulate how dedicated CAV lane operations can be implemented without significant change in existing infrastructure.

Today, automated vehicles mostly rely on ego vehicle sensors such as cameras, radar or LiDAR sensors that are limited in their sensing capability and range. Vehicle-to-everything (V2X) communication has the potential to appropriately complement these sensors and even allow for a cooperative, proactive interaction of vehicles. As such, V2X communication might play a vital role on the way to smart and efficient traffic solutions. In the public funded research project UK Autodrive, we are currently investigating and experimentally evaluating V2X-based applications based on dedicated short range communication (DSRC). Moreover, the novel application intersection priority management (IPM) is part of the research project. IPM aims at automating intersections in such a way that vehicles can pass safely and even more efficiently without the use of traffic lights or signs.

In the last years various advanced driver assistance systems (ADAS) have been introduced on the market. More highly advanced functions up to automated driving functions are currently under research. By means of these functions partly automated driving in specific situations is already or will be realized soon, e.g. traffic jam assist. Besides the technical challenges to develop such automated driving functions for complex situations, e.g. construction or intersection areas, new approaches for the evaluation of these functions under different driving conditions are necessary, in order to assess the benefits and identify potential weaknesses. Classical approaches for evaluation and market sign off will require an extensive testing, which results in high costs and time demands. Therefore the classical approaches are hardly feasible taking into account higher levels of support and automation. Today the final sign-off requires a high amount of real world tests.

This paper presents experimental results that validate eco-driving and eco-heating strategies developed for connected and automated vehicles (CAVs). By exploiting vehicle-to-infrastructure (V2I) communications, traffic signal timing, and queue length estimations, optimized and smoothed speed profiles for the ego-vehicle are generated to reduce energy consumption. Next, the planned eco-trajectories are incorporated into a real-time predictive optimization framework that coordinates the cabin thermal load (in cold weather) with the speed preview, i.e., eco-heating. To enable eco-heating, the engine coolant (as the only heat source for cabin heating) and the cabin air are leveraged as two thermal energy storages. Our eco-heating strategy stores thermal energy in the engine coolant and cabin air while the vehicle is driving at high speeds, and releases the stored energy slowly during the vehicle stops for cabin heating without forcing the engine to idle to provide the heating source.

By utilizing the vehicle to infrastructure communication, the conventional Green Light Optimized Speed Advisory (GLOSA) applications give speed advisory range for drivers to travel to pass at the green light. However, these systems do not consider the traffic between the ego vehicle and the traffic light location, resulting in inaccurate speed advisories. Therefore, the driver needs to intuitively adjust the vehicle's speed to pass at the green light and avoid traffic in these scenarios. Furthermore, inaccurate speed advisories may result in unnecessary acceleration and deceleration, resulting in poor fuel efficiency and comfort. To address these shortcomings of conventional GLOSA, in this study, we proposed the utilization of collaborative perception messages shared by smart infrastructures to create an enhanced speed advisory for the connected vehicle drivers and automated vehicles.

Connectivity in ground vehicles allows vehicles to share crucial vehicle data, such as vehicle acceleration and speed, with each other. Using sensors such as radars and lidars, on the other hand, the intravehicular distance between a leader vehicle and a host vehicle can be detected. Cooperative Adaptive Cruise Control (CACC) builds upon ground vehicle connectivity and sensor information to form convoys with automated car following. CACC can also be used to improve fuel economy and mobility performance of vehicles in the said convoy. In this paper, a CACC system is presented, where the acceleration of the lead vehicle is used in the calculation of desired vehicle speed. In addition to the smooth car following abilities, the proposed CACC also has the capability to calculate a speed profile for the ego vehicle that is fuel efficient, making it an Ecological CACC (Eco-CACC) model.

In this work, we outline a process for traffic light detection in the context of autonomous vehicles and driver assistance technology features. For our approach, we leverage the automatic annotations from virtually generated data of road scenes. Using the automatically generated bounding boxes around the illuminated traffic lights themselves, we trained an 8-layer deep neural network, without pre-training, for classification of traffic light signals (green, amber, red). After training on virtual data, we tested the network on real world data collected from a forward facing camera on a vehicle. Our new region proposal technique uses color space conversion and contour extraction to identify candidate regions to feed to the deep neural network classifier. Depending on time of day, we convert our RGB images in order to more accurately extract the appropriate regions of interest and filter them based on color, shape and size. These candidate regions are fed to a deep neural network.