Search

Autonomous driving perceives surroundings with line-of-sight sensors that are compromised under environmental uncertainties. To achieve real time global information in high definition map, we investigate to share perception information among connected and automated vehicles. However, it is challenging to achieve real time perception sharing under varying network dynamics in automotive edge computing. In this paper, we propose a novel real time dynamic map, named LiveMap to detect, match, and track objects on the road. We design the data plane of LiveMap to efficiently process individual vehicle data with multiple sequential computation components, including detection, projection, extraction, matching and combination. We design the control plane of LiveMap to achieve adaptive vehicular offloading with two new algorithms (central and distributed) to balance the latency and coverage performance based on deep reinforcement learning techniques. We conduct extensive evaluation through both realistic experiments on a small-scale physical testbed and network simulations on an edge network simulator. The results suggest that LiveMap significantly outperforms existing solutions in terms of latency, coverage, and accuracy., Comment: This paper is accepted by IEEE Transactions on Intelligent Vehicles

Today very few deep learning-based mobile augmented reality (MAR) applications are applied in mobile devices because they are significantly energy-guzzling. In this paper, we design an edge-based energy-aware MAR system that enables MAR devices to dynamically change their configurations, such as CPU frequency, computation model size, and image offloading frequency based on user preferences, camera sampling rates, and available radio resources. Our proposed dynamic MAR configuration adaptations can minimize the per frame energy consumption of multiple MAR clients without degrading their preferred MAR performance metrics, such as latency and detection accuracy. To thoroughly analyze the interactions among MAR configurations, user preferences, camera sampling rate, and energy consumption, we propose, to the best of our knowledge, the first comprehensive analytical energy model for MAR devices. Based on the proposed analytical model, we design a LEAF optimization algorithm to guide the MAR configuration adaptation and server radio resource allocation. An image offloading frequency orchestrator, coordinating with the LEAF, is developed to adaptively regulate the edge-based object detection invocations and to further improve the energy efficiency of MAR devices. Extensive evaluations are conducted to validate the performance of the proposed analytical model and algorithms., Comment: This is a personal copy of the authors. Not for redistribution. The final version of this paper was accepted by IEEE Transactions on Mobile Computing

Autonomous driving needs various line-of-sight sensors to perceive surroundings that could be impaired under diverse environment uncertainties such as visual occlusion and extreme weather. To improve driving safety, we explore to wirelessly share perception information among connected vehicles within automotive edge computing networks. Sharing massive perception data in real time, however, is challenging under dynamic networking conditions and varying computation workloads. In this paper, we propose LiveMap, a real-time dynamic map, that detects, matches, and tracks objects on the road with crowdsourcing data from connected vehicles in sub-second. We develop the data plane of LiveMap that efficiently processes individual vehicle data with object detection, projection, feature extraction, object matching, and effectively integrates objects from multiple vehicles with object combination. We design the control plane of LiveMap that allows adaptive offloading of vehicle computations, and develop an intelligent vehicle scheduling and offloading algorithm to reduce the offloading latency of vehicles based on deep reinforcement learning (DRL) techniques. We implement LiveMap on a small-scale testbed and develop a large-scale network simulator. We evaluate the performance of LiveMap with both experiments and simulations, and the results show LiveMap reduces 34.1% average latency than the baseline solution., Comment: This paper is accepted by INFOCOM 2021

Applying deep learning to object detection provides the capability to accurately detect and classify complex objects in the real world. However, currently, few mobile applications use deep learning because such technology is computation-intensive and energy-consuming. This paper, to the best of our knowledge, presents the first detailed experimental study of a mobile augmented reality (AR) client's energy consumption and the detection latency of executing Convolutional Neural Networks (CNN) based object detection, either locally on the smartphone or remotely on an edge server. In order to accurately measure the energy consumption on the smartphone and obtain the breakdown of energy consumed by each phase of the object detection processing pipeline, we propose a new measurement strategy. Our detailed measurements refine the energy analysis of mobile AR clients and reveal several interesting perspectives regarding the energy consumption of executing CNN-based object detection. Furthermore, several insights and research opportunities are proposed based on our experimental results. These findings from our experimental study will guide the design of energy-efficient processing pipeline of CNN-based object detection., Comment: This is a personal copy of the authors. Not for redistribution. The final version of this paper was accepted by IEEE Transactions on Green Communications and Networking

As vehicles playing an increasingly important role in people's daily life, requirements on safer and more comfortable driving experience have arisen. Connected vehicles (CVs) can provide enabling technologies to realize these requirements and have attracted widespread attentions from both academia and industry. These requirements ask for a well-designed computing architecture to support the Quality-of-Service (QoS) of CV applications. Computation offloading techniques, such as cloud, edge, and fog computing, can help CVs process computation-intensive and large-scale computing tasks. Additionally, different cloud/edge/fog computing architectures are suitable for supporting different types of CV applications with highly different QoS requirements, which demonstrates the importance of the computing architecture design. However, most of the existing surveys on cloud/edge/fog computing for CVs overlook the computing architecture design, where they (i) only focus on one specific computing architecture and (ii) lack discussions on benefits, research challenges, and system requirements of different architectural alternatives. In this paper, we provide a comprehensive survey on different architectural design alternatives based on cloud/edge/fog computing for CVs. The contributions of this paper are: (i) providing a comprehensive literature survey on existing proposed architectural design alternatives based on cloud/edge/fog computing for CVs, (ii) proposing a new classification of computing architectures based on cloud/edge/fog computing for CVs: computation-aided and computation-enabled architectures, (iii) presenting a holistic comparison among different cloud/edge/fog computing architectures for CVs based on functional requirements of CV systems, including advantages, disadvantages, and research challenges., Comment: This is a personal copy of the authors. Not for redistribution. The final version of this paper is available through the IEEE Xplore Digital Library, at the link: https://ieeexplore.ieee.org/document/9184917, with the DOI: 10.1109/COMST.2020.3020854

We study the problem of policy repair for learning-based control policies in safety-critical settings. We consider an architecture where a high-performance learning-based control policy (e.g. one trained as a neural network) is paired with a model-based safety controller. The safety controller is endowed with the abilities to predict whether the trained policy will lead the system to an unsafe state, and take over control when necessary. While this architecture can provide added safety assurances, intermittent and frequent switching between the trained policy and the safety controller can result in undesirable behaviors and reduced performance. We propose to reduce or even eliminate control switching by `repairing' the trained policy based on runtime data produced by the safety controller in a way that deviates minimally from the original policy. The key idea behind our approach is the formulation of a trajectory optimization problem that allows the joint reasoning of policy update and safety constraints. Experimental results demonstrate that our approach is effective even when the system model in the safety controller is unknown and only approximated.

As autonomous vehicles have benefited the society, understanding the dynamic change of humans' trust during human-autonomous vehicle interaction can help to improve the safety and performance of autonomous driving. We designed and conducted a human subjects study involving 19 participants. Each participant was asked to enter their trust level in a Likert scale in real-time during experiments on a driving simulator. We also collected physiological data (e.g., heart rate, pupil size) of participants as complementary indicators of trust. We used analysis of variance (ANOVA) and Signal Temporal Logic (STL) to analyze the experimental data. Our results show the influence of different factors (e.g., automation alarms, weather conditions) on trust, and the individual variability in human reaction time and trust change.

Deep Neural Networks (DNNs) have tremendous potential in advancing the vision for self-driving cars. However, the security of DNN models in this context leads to major safety implications and needs to be better understood. We consider the case study of steering angle prediction from camera images, using the dataset from the 2014 Udacity challenge. We demonstrate for the first time adversarial testing-time attacks for this application for both classification and regression settings. We show that minor modifications to the camera image (an L2 distance of 0.82 for one of the considered models) result in mis-classification of an image to any class of attacker's choice. Furthermore, our regression attack results in a significant increase in Mean Square Error (MSE) by a factor of 69 in the worst case., Comment: Preprint of the work accepted for publication at the IEEE Workshop on the Internet of Safe Things, San Francisco, CA, USA, May 23, 2019

Connected and automated vehicle (CAV) technology is one of the promising solutions to addressing the safety, mobility and sustainability issues of our current transportation systems. Specifically, the control algorithm plays an important role in a CAV system, since it executes the commands generated by former steps, such as communication, perception, and planning. In this study, we propose a consensus algorithm to control the longitudinal motion of CAVs in real time. Different from previous studies in this field where control gains of the consensus algorithm are pre-determined and fixed, we develop algorithms to build up a lookup table, searching for the ideal control gains with respect to different initial conditions of CAVs in real time. Numerical simulation shows that, the proposed lookup table-based consensus algorithm outperforms the authors' previous work, as well as van Arem's linear feedback-based longitudinal motion control algorithm in all four different scenarios with various initial conditions of CAVs, in terms of convergence time and maximum jerk of the simulation run., Comment: 2019 American Control Conference (ACC)Philadelphia, PA, USA, July 10-12, 2019978-1-5386-7928-9

Agent-based modeling and simulation (ABMS) has been a popular approach to modeling autonomous and interacting agents in a multi-agent system. Specifically, ABMS can be applied to connected and automated vehicles (CAVs), since CAVs can be driven autonomously with the help of on-board sensors, and cooperate with each other through vehicle-to-everything (V2X) communications. In this work, we apply ABMS to CAVs using the game engine Unity3D, taking advantage of its visualization capability and other capabilities. Agent-based models of CAVs are built in the Unity3D environment, where vehicles are enabled with connectivity and autonomy by C#-based scripting API. We also build a simulation network in Unity3D based on the city of Mountain View, California. A case study of cooperative on-ramp merging has been carried out with the proposed distributed consensus-based protocol, and then compared with the human-in-the-loop simulation where the on-ramp vehicle is driven by four different human drivers on a driving simulator. The benefits of introducing the proposed protocol are evaluated in terms of travel time, energy consumption, and pollutant emissions. It is shown from the results that the proposed cooperative on-ramp merging protocol can reduce average travel time by 7%, reduce energy consumption and pollutant emissions by 8% and 58%, respectively, and guarantee the driving safety when compared to the human-in-the-loop scenario., Comment: 14 pages, 6 figures. 2019 Transportation Research Board Annual Meeting

Distributed controllers are often necessary for a multi-agent system to satisfy safety properties such as collision avoidance. Communication and coordination are key requirements in the implementation of a distributed control protocol, but maintaining an all-to-all communication topology is unreasonable and not always necessary. Given a safety objective and a controller implementation, we consider the problem of identifying when agents need to communicate with one another and coordinate their actions to satisfy the safety constraint. We define a coordination-free controllable predecessor operator that is used to derive a subset of the state space that allows agents to act independently, without consulting other agents to double check that the action is safe. Applications are shown for identifying an upper bound on connection delays and a self-triggered coordination scheme. Examples are provided which showcase the potential for designers to visually interpret a system's ability to tolerate delays when initializing a network connection., Comment: In Proceedings SCAV 2018, arXiv:1804.03406