For decades, researchers on Vehicular Ad-hoc Networks (VANETs) and autonomous vehicles presented various solutions for vehicular safety and autonomy, respectively. Yet, the developed work in these two areas has been mostly conducted in their own separate worlds, and barely affect one-another despite the obvious relationships. In the coming years, the Internet of Vehicles (IoV), encompassing sensing, communications, connectivity, processing, networking, and computation is expected to bridge many technologies to offer value-added information for the navigation of self-driving vehicles, to reduce vehicle on board computation, and to deliver desired functionalities. Potentials for bridging the gap between these two worlds and creating synergies of these two technologies have recently started to attract significant attention of many companies and government agencies. In this article, we first present a comprehensive survey and an overview of the emerging key challenges related to the two worlds of Vehicular Clouds (VCs) including communications, networking, traffic modelling, medium access, VC Computing (VCC), VC collation strategies, security issues, and autonomous driving (AD) including 3D environment learning approaches and AD enabling deep-learning, computer vision and Artificial Intelligence (AI) techniques. We then discuss the recent related work and potential trends on merging these two worlds in order to enrich vehicle cognition of its surroundings, and enable safer and more informed and coordinated AD systems. Compared to other survey papers, this work offers more detailed summaries of the most relevant VCs and ADs systems in the literature, along with some key challenges and insights on how different technologies fit together to deliver safety, autonomy and infotainment services.

This paper presents a novel Advanced Activity-Aware (AAA) scheme to optimize and improve Multi-Channel Operations based on the IEEE 1609.4 standard in wireless access vehicular environments (WAVE). The proposed scheme relies on the awareness of the vehicular safety load to dynamically find an optimal setup for switching between service channel intervals (SCHI) and control channel intervals (CCHI). SCHI are chosen for non-critical applications (e.g. infotainment), while CCHI are utilized for critical applications (e.g. safety-related). We maximize the channel utilization without sacrificing other application requirements such as latency and bandwidth. Our scheme is implemented and evaluated network simulator-3 (NS3). We guarantee the default Synchronization Interval (SI), like implemented by the standard in vehicular ad hoc networks (VANETs), when tested on real-time simulations of vehicular cloud (VC) load and VANET setups using NS3. We also evaluate a Markov Decision Process (MDP) based scheme and a fast greedy heuristic to optimize the problem of vehicular task placement with both IEEE 1609.4 and an opportunistic V2I version of the proposed AAA scheme. Our solution offers the reward of the VC by taking into account the overall utilization of the distributed virtualized VCs resources and vehicular bag-of-tasks (BOTs) placements both sequentially and in parallel. We present the simulation metrics proving that our proposed solution significantly improve the throughput and decreases the average delay of uploaded packets used for non-safety applications, while maintaining reliable communication (via CCHI) for safety-related applications similar to the IEEE 1609.4 standard.

In this paper, we present a novel approach to track recognized 3D vehicle point clouds from LIDAR scans. This technique is based on tracing the 3D anchor boxes of these vehicles, recognized though convolutional neural networks (CNNs). Exploiting the 3D CNNs detection of vehicles and persons, and the Extended Kalman Filters (EKF) two-steps process for prediction and update, the proposed scheme guarantees the awareness of moving detected objects and improves the perception of Autonomous Vehicles (AV). The proposed scheme reduces the usage of the expensive detection process of feeding the point cloud to CNN by tracking the 3D rectangular coordinates containing already detected objects from early Velodyne scans of the driving sequences. The testing of the proposed method on the well-known KITTI dataset, featuring LIDAR scans of realistic vehicular environments. Results show the merits of the proposed scheme in achieving high tracking accuracy.

In the past two years, calls for developing synergistic links between the two worlds of vehicular ad-hoc networks (VANETs) and autonomous vehicles have significantly gone up to achieve further on-road safety and benefits for end-users. In this paper, we present our vision to create such a beneficial link by designing a multimodal scheme for object detection, recognition, and mapping based on the fusion of stereo camera frames, point cloud Velodyne LIDAR scans, and vehicle-to-vehicle (V2V) basic safety messages (BSMs) exchanges using VANET protocols. Exploiting the high similarities in the underlying manifold properties of the three data sets, and their high neighborhood correlation, the proposed scheme employs semi-supervised manifold alignment to merge the key features of rich texture descriptions of objects from 2-D images, depth and distance between objects provided by 3-D point cloud, and the awareness of self-declared vehicles from BSMs' 3-D information including the ones not seen by camera and LIDAR. The proposed scheme is applied to create joint pixel-to-point-cloud and pixel-to-V2V correspondences of objects in frames from the KITTI Vision Benchmark Suite, using a semi-supervised manifold alignment, to achieve camera-LIDAR and camera-V2V mapping of their recognized objects. We present the alignment accuracy results over two different driving sequences and show the additional acquired knowledge of objects from the various input modalities. We also study the effect of the number of neighbors employed in the alignment process on the alignment accuracy. With proper choice of parameters, the testing of our proposed scheme over two entire driving sequences exhibits 100% accuracy in the majority of cases, 74%-92% and 50%-72% average alignment accuracy for vehicles and pedestrians and up to 150% additional object recognition of the testing vehicle's surrounding.

In this paper, we design a multimodal framework for object detection, recognition and mapping based on the fusion of stereo camera frames, point cloud Velodyne LIDAR scans, and Vehicle-to-Vehicle (V2V) Basic Safety Messages (BSMs) that are exchanged using Dedicated Short Range Communication (DSRC). We merge the key features of rich texture descriptions of objects from 2D images using Convolutional Neural Networks (CNN). In addition, depth and distance between objects are provided by the 3D LIDAR point cloud and the awareness of hidden vehicles is achieved from BSMs' beacons. We present a joint pixel to point cloud and pixel to V2V correspondence of objects in frames of driving sequences in the KITTI Vision Benchmark Suite. We achieve this by using a semi-supervised manifold alignment approach to achieve camera-LIDAR and camera-V2V mapping of their recognized persons and cars that have the same underlying manifold.

In this paper, we consider the use of CUDAbased Graphics Processing Units (GPUs) as high performance parallel computing for the purpose of accelerating the application task scheduling in Vehicular Cloud Computing (VCC) systems. We leverage the Single Instruction Multiple Data (SIMD) mode in GeneralPurpose Graphic Processing Units (GPGPUs) to solve the value iteration algorithm of the defined Markov Decision Process (MDP) of task scheduling on real VCC. We consider opportunistically available Vehicle to Infrastructure (V2I) communication in Dedicated Short Range Communication (DSRC) used in Vehicular ad hoc networks (VANETs) for the vehicular clouds.

In this paper, we investigate the performance analysis of dual hop relaying system consisting of asymmetric Radio Frequency (RF)/Free Optical Space (FSO) links. The RF channels follow a Rayleigh distribution and the optical links are subject to Gamma-Gamma fading. We also introduce impairments to our model and we suggest Partial Relay Selection (PRS) protocol with Amplify-and-Forward (AF) fixed gain relaying. The benefits of employing optical communication with RF, is to increase the system transfer rate and thus improving the system bandwidth. Many previous research attempts assuming ideal hardware (source, relays, etc.) without impairments. In fact, this assumption is still valid for low-rate systems. However, these hardware impairments can no longer be neglected for high-rate systems in order to get consistent results. Novel analytical expressions of outage probability and ergodic capacity of our model are derived taking into account ideal and non-ideal hardware cases. Furthermore, we study the dependence of the outage probability and the system capacity considering, the effect of the correlation between the outdated CSI (Channel State Information) and the current source-relay link, the number of relays, the rank of the selected relay and the average optical Signal to Noise Ratio (SNR) over weak and strong atmospheric turbulence. We also demonstrate that for a non-ideal case, the end-to-end Signal to Noise plus Distortion Ratio (SNDR) has a certain ceiling for high SNR range. However, the SNDR grows infinitely for the ideal case and the ceiling caused by impairments no longer exists. Finally, numerical and simulation results are presented.

The Dedicated Short Range Communication (DSRC) technology has been used in Vehicular communication to enable short-lived safety and non-safety applications based on Vehicle to Vehicle (V2V) communications over the Control Channel (CCH) and Vehicle to Infrastructure (V2I) communications over the Service Channels (SCHs) in Vehicular Ad hoc Networks (VANETs). With the under-utilized advanced computation, communication and storage resources in On-Board Units (OBUs) of modern vehicles, Vehicular Clouds (VCs) are used to manage coalitions of affordable resources in Vehicles in order to host infotainment applications used by other vehicles on the move. In this paper, we introduce an Advanced Activity-Aware (AAA) scheme for Multi-Channel Operations based on 1609.4 in MAC Protocol in Wireless Access in Vehicular Environments (WAVE). The AAA aims at dynamically achieving an optimal setup of Control Channel Interval (CCHI) and Service Channel Interval (SCHI) by reducing the inactivity interval while maintaining a default Synchronization Interval (SI) between all vehicles. We evaluate the performance of our proposed scheme through real-time simulation of vehicular cloud load and VANET communications using NS3. The simulation results indicate that our proposed scheme increases significantly the throughput and reduces the average delay of uploaded packets of non-safety applications from the VC to the Road Side Unit (RSU) while maintaining a V2V communication for safety similar to that of the 1609.4 standard.