Abstract:Future vehicles will become smarter and more connected. With the merging and sharing of data generated by onboard sensors, next-generation vehicular applications (e.g., cooperative lane change) are emerging that create a significant demand for sufficient computing resources to conduct time-critical and data-intensive tasks. Owing to the space, weight, and cost constraints, the computing capacity of most vehicles may not be high enough to handle such tasks. On the other hand, offloading these tasks to the cloud is not applicable, due to the considerable transmission delay. A new computing paradigm known as vehicular fog computing (VFC) has been proposed that pushes computing and communication resources to the edge of the network. Its key idea is to offload computational tasks from the client vehicles to fog nodes located for example at cellular base stations or buses with extra computing power. In VFC, only one-hop communication is required, which greatly shortens the transmission delay. However, due to the mobility of vehicles, the density of client vehicles and therefore the amount of tasks generated by them vary spatiotemporally. Meanwhile, the availability of fog nodes carried by vehicles, called vehicular fog nodes, depends on the driving routes of the carriers. The spatiotemporal variation in both the supply and demand of computing services adds a layer of complexity to enable reliable VFC-based services. In this dissertation, the focus is on designing task allocation and resource scheduling algorithms via various mathematical models to enable high-quality and low-latency VFC-based services for vehicular applications. The feasibility and challenges of applying VFC for real-time analytics of a crowdsourced dash camera video is investigated. Furthermore, a framework for latency and quality optimized task allocation in VFC is presented and a task-offloading framework for visual-based assisted driving is proposed. Finally, the design of a QoI and latency aware task allocation scheme for vehicle-based visual crowdsourcing is presented, which takes into account vehicle mobility and the spatiotemporal variation in the workload of vehicular fog nodes. Analytical studies were conducted in order to answer the research questions and evaluate the effectiveness of the designed algorithms, using real-world application profiles and traffic data as input.

FlexSensing: A QoI and Latency-Aware Task Allocation Scheme for Vehicle-Based Visual Crowdsourcing Via Deep Q-Network.

Task Allocation and Resource Scheduling in Vehicular Fog Computing

Toward optimal allocation of location dependent tasks in crowdsensing

Hybrid Sensor Network with Edge Computing for AI Applications of Connected Vehicles

Adaptive Task Offloading in Vehicular Edge Computing Networks: a Reinforcement Learning Based Scheme

Fog Following Me: Latency and Quality Balanced Task Allocation in Vehicular Fog Computing

An End-to-End Smart Predict-then-Optimize Framework for Vehicle Relocation Problems in Large-Scale Vehicle Crowd Sensing

Chameleon: Latency and Resolution Aware Task Offloading for Visual-Based Assisted Driving

Delay Optimization in Vehicular Edge Computing with Sensing Information Fusion and Heterogeneous Tasks

QUEST: Quality-informed Multi-agent Dispatching System for Optimal Mobile Crowdsensing

A deep reinforcement learning assisted task offloading and resource allocation approach towards self-driving object detection

Ilocus: Incentivizing Vehicle Mobility to Optimize Sensing Distribution in Crowd Sensing

High and Low Resolution Tradeoffs in Roadside Multimodal Sensing

QoS and Fairness Oriented Dynamic Computation Offloading in the Internet of Vehicles based on Estimate Time of Arrival

Evenness-Aware Data Collection for Edge-Assisted Mobile Crowdsensing in Internet of Vehicles

Cloud-Edge–End Collaborative Task Offloading in Vehicular Edge Networks: A Multilayer Deep Reinforcement Learning Approach

Cosense

High-resolution large-scale urban traffic speed estimation with multi-source crowd sensing data

Edge-Assisted Accelerated Cooperative Sensing for CAVs: Task Placement and Resource Allocation

Exploring the sensing power of mixed vehicle fleets

Sustainable Road Pothole Detection: A Crowdsourcing Based Multi-Sensors Fusion Approach