Featured Articles of the VTS Section in IEEE Access
Reach over 8 million monthly users on IEEE Xplore® by publishing your vehicular technology research in the IEEE Vehicular Technology (VTS) Section of IEEE Access. Benefit from a rapid, high-quality peer review process of just 4-6 weeks.
Take a look at some of the highly cited articles in the VTS Section below, addressing key topics in vehicular technology.
FRENETIX: A High-Performance and Modular Motion Planning Framework for Autonomous Driving
Authors: R. Trauth; K. Moller; G. Würsching; J. Betz
Abstract: This paper presents FRENETIX, an open-source, high-performance, and modular motion planning framework designed to meet the rigorous demands of real-world autonomous driving. Built on a sampling-based trajectory planning algorithm, FRENETIX excels in navigating both static and dynamic environments by prioritizing safety, comfort, and path precision through multi-objective optimization.
User Plane Function (UPF) Allocation for C-V2X Network Using Deep Reinforcement Learning
Authors: P. Sasithong; T. Sanguanpuak; P. Vanichchanunt; L. Wuttisittikulkij
Abstract: This paper presents a novel online learning approach for optimizing the placement of User Plane Functions (UPFs) in Cellular Vehicle-to-Everything networks using Deep Reinforcement Learning techniques. By leveraging Deep Q-Network and Actor-Critic algorithms, both approaches can reduce computation time and end-to-end communication latency significantly, up to 40% compared to the baseline approaches.
XAI-ADS: An Explainable Artificial Intelligence Framework for Enhancing Anomaly Detection in Autonomous Driving Systems
Authors: S. Nazat; L. Li; M. Abdallah
Abstract: This paper proposes an end-to-end explainable AI (XAI) framework to interpret the decision-making of AI models for anomaly detection of autonomous vehicles (AVs). It provides both global and local explanations for anomaly detection AI models on AVs. It also proposes two novel XAI-based feature selection approaches for identifying contributions of significant features for anomaly classification of an AV. The framework is benchmarked using six different AI models on two autonomous driving datasets with different characteristics.
Downsizing of Electric Motors by Utilizing Differential Gear for Traction and Hydraulic Pump of Heavy Mobile Machine
Authors: T. Tyni; J. Vepsäläinen
Abstract: To expedite the expensive electrification process and transition to more energy-efficient technologies, the heavy mobile machinery industry needs comprehensive system-level solutions that maximize component utilization. A novel utilization of a differential gear mechanism is proposed to couple two electric motors and a hydraulic pump. The analysis employed simulations to examine simultaneous operation of traction drive and working hydraulics in a wheel loader application. It was shown that the proposed system can significantly reduce the torque demand of the traction electric motor in loading scenarios.