Optimization of Micromobility Integration in IoV Systems: Energy-Efficient Management Through Mobile RSUs and Machine-Learning Based Approaches

Micromobility, encompassing low-speed transport modes such as e-scooters and bicycles, plays a pivotal role in enhancing urban mobility within the Intelligent Transportation System (ITS). For such an environment, a multitude of challenges arises, including optimizing system architecture and managing energy consumption efficiently. This paper presents an innovative approach aimed at improving micromobility integration within an Internet of Vehicles (IoV) ecosystem by leveraging Vehicular Adhoc Networks (VANETs) and IEEE 802.11p communication standard. To reduce energy consumption while maintaining costeffective infrastructure, we propose an energy-saving strategy that optimizes the operation by introducing mobile RSUs (mRSUs) with dynamic and adaptive states called (”ON/OFF”). Furthermore, we introduce a k-Nearest Neighbors (KNN) algorithm, leveraging Machine Learning (ML), to predict the optimal speed of these mRSUs, enhancing their efficiency and adaptability. The effectiveness of these approaches is rigorously evaluated based on the activation time of the mRSU, the total energy consumption metric, and using a meta-analysis method.