AI-Driven Energy Harvesting Aware Path Planning for Electric Vehicles under Stochastic Road Vibrations: A review

Abstract

Although electric vehicles (EVs) are playing an important role in environmentally sustainable transport, there are still significant challenges to be addressed, such as the range and energy efficiency of EVs. Conventional approaches to EV path planning are primarily concerned with optimizing travel time or distance without considering the effect of uncertain road conditions and energy consumption. In particular, road-induced vibrations are typically regarded as unavoidable energy losses because they always excite the suspension systems of vehicles. In the context of stochastic road vibrations, this paper presents a comprehensive review of AI-based electric vehicle path planning with consideration of energy harvesting. After a brief discussion of the characteristics of electric vehicle engines and energy consumption modeling, the paper explores the reasons for road-induced vibrations and stochastic modeling approaches. The paper further investigates the mechanisms of vibration energy harvesting and their integration with EVs. In this paper, we discuss energy-efficient path planning from the perspective of recent advances in AI techniques such as ML, RN, and graph neural networks. Taking into account practical constraints such as travel time, comfort, and battery life, this review work explores integrated frameworks that can simultaneously achieve maximum vibration energy harvesting while minimizing battery energy consumption. The scalability of the proposed framework to networks of connected and autonomous EVs, existing studies, and challenges associated with uncertainty, data availability, and real-time processing are all discussed in this paper. The paper concludes by discussing possible research directions that can assist in developing EV navigation systems that are intelligent and energy-efficient.