Scanning the Issue

Azim Eskandarian
2021 IEEE transactions on intelligent transportation systems (Print)  
Research Advances and Challenges of Autonomous and Connected Ground Vehicles A. Eskandarian, C. Wu, and C. Sun This article introduces a representative architecture of connected autonomous vehicles (CAVs) and surveys the latest research advances, methods, and algorithms for sensing, perception, planning, and control of CAVs. It reviews the state-of-the-art and state-of-the-practice (when applicable) of a multi-layer perception-planning-control architecture including on-board sensors and
more » ... r communications, the methods of sensor fusion and localization and mapping in the perception layer, the algorithms of decision making and trajectory planning in the planning layer, and the control strategies of trajectory tracking in the control layer. Furthermore, the implementations and impact of vehicle connectivity and the corresponding consequential challenges of cooperative perception, complex connected decision making, and multivehicle controls are summarized and their significant research issues enumerated. Deep learning has emerged as a prominent class of techniques for autonomous driving. This article reviews the stateof-the-art techniques using deep neural networks to control autonomous vehicles. The strengths and limitations of current techniques are identified through comparative analysis. Important advancements in the field are discussed, current trends are summarized, and future prospects are reviewed. Furthermore, research challenges, which need to be solved before these algorithms are ready for safe deployment in the real world, are discussed and potential solutions are identified. Finally, recommendations for future research directions in the field are given. Presented in this article is a novel method for the mapping and semantic modeling of an underground parking lot using 3D point clouds collected by a low-cost backpack laser scanning (BLS) or LiDAR system. Their method consists of two parts: a simultaneous localization and mapping (SLAM) algorithm based on sparse point clouds (SPC) and a semantic modeling algorithm based on a modified PointNet model. The main contributions of this article are as follows: 1) a probability frontend framework for the alignment of point clouds using the local point cloud surface variance Digital Object Identifier 10.1109/TITS.2021.3052540 as the weight of registration, which modifies registration failure caused by the lack of features in sparse point clouds; 2) a robust submap-based strategy for loop closure detection and back-end optimization under sparse point clouds; and 3) a modified PointNet model for classifying the point clouds of underground parking lots into four categories: ceiling, floor, wall, others. The experimental results show that their SPC-SLAM algorithm achieves cm-level accuracy (0.09% trajectory error rate) after closed-loop processing in a global navigation satellite system (GNSS)-denied underground parking lot, and precision of 84.8% in semantic segmentation. An integer program with the objective of maximizing the police visibility rate and the additional constraint of response time guarantee is formulated to model the cityscale patrolling (CSP) problem. The original CSP is decomposed into two weakly-coupled subproblems, minimizing police problem (MinP) and maximizing PVR (MaxP) problem. By exploiting the subproblem structures, a polynomial time approximation algorithm is proposed for MinP, and a polynomial time optimal algorithm is proposed for MaxP. The authors prove that such a decomposition can provide the 1-α approximation ratio, where α is the percentage of the police used in MinP. To further improve patrolling efficiency, a grafting mechanism is proposed to integrate the two subproblems' solutions. Vehicle logo recognition provides an important supplement to vehicle make and model analysis. This article presents a cascaded deep convolutional network for directly recognizing vehicle logos without depending on the existence of license plates. This is a two-stage processing framework composed of a region proposal network and a convolutional capsule network. First, potential region proposals that might contain vehicle logos are generated by the region proposal network. Then, the convolutional capsule network classifies these region proposals into the background and different types of vehicle logos. The proposed method performs effectively and robustly in recognizing vehicle logos of various conditions. Driver Intervention Detection via Real-Time Transfer Function Estimation W. S. Schinkel, T. P. J. van der Sande, and H. Nijmeijer A novel driver intervention detection method for automated vehicles is presented and tested. The transfer function between 1558-0016
doi:10.1109/tits.2021.3052540 fatcat:wvccn3i32jdaxoov6mibk2tlku