Autonomous driving economic car-following motion strategy based on adaptive rollout model-based policy optimization

Abstract

Reinforcement learning (RL) is a powerful framework with significant potential to enhance autonomous driving (AD) performance. However, its trial-and-error nature presents significant hurdles in terms of safety, efficiency, and stability. To address these challenges, we propose an adaptive rollout model-based policy optimization (AR-MBPO) algorithm tailored for car-following motion planning in autonomous electric vehicles (AEVs). The algorithm improves overall performance by incorporating a error-aware ensemble environment model and leveraging branched rollouts for efficient sample collection and policy optimization. A key innovation of AR-MBPO is an adaptive rollout mechanism that dynamically adjusts based on predictive accuracy, mitigating the impact of model inaccuracies. Additionally, energy efficiency is explicitly integrated into the optimization process to minimize energy consumption. We evaluate AR-MBPO through AEV car-following simulations, where it demonstrates superior performance, including rapid convergence, and reduced reliance on real-world interactions. The method simultaneously optimizes safety, traffic efficiency, and energy efficiency through dynamic distance adjustments, as evidenced by a 0% collision rate in testing scenarios and an 8.2% energy consumption reduction compared to non-energy-aware baselines. The results suggest potential applications in AD systems for improved safety and energy efficiency.