Energy-Aware Path Following: Comparative Analysis of Reinforcement Learning and NMPC for Electric Vehicles

TL;DR

This paper compares reinforcement learning and nonlinear model predictive control for energy-aware path following in electric vehicles. The results show that proximal policy optimization can be transferred to unseen tracks and vehicle models. The study demonstrates the potential of reinforcement learning for energy-efficient control.

Why It Matters

The research has practical significance for improving the driving range of electric vehicles through energy-efficient control strategies. It can also contribute to the development of autonomous vehicles with optimized energy consumption. The findings can be applied to various vehicle models and tracks, making it a valuable contribution to the field.

Key Findings

• comparative analysis of reinforcement learning and nonlinear model predictive control
• transferability of proximal policy optimization to unseen tracks and vehicle models
• energy efficiency of reinforcement learning for electric vehicles