This paper proposes a novel inverse reinforcement learning framework using a diffusion-based adaptive lookahead planner (IRL-DAL) for autonomous vehicles. Training begins with imitation from an expert finite state machine (FSM) controller to provide a stable initialization. Environment terms are combined with an IRL discriminator signal to align with expert goals. Reinforcement learning (RL) is then performed with a hybrid reward that combines diffuse environmental feedback and targeted IRL rewards. A conditional diffusion model, which acts as a safety supervisor, plans safe paths. It stays in its lane, avoids obstacles, and moves smoothly. Then, a learnable adaptive mask (LAM) improves perception. It shifts visual attention based on vehicle speed and nearby hazards. After FSM-based imitation, the policy is fine-tuned with Proximal Policy Optimization (PPO). Training is run in the Webots simulator with a two-stage curriculum. A 96\% success rate is reached, and collisions are reduced to 0.05 per 1k steps, marking a new benchmark for safe navigation. By applying the proposed approach, the agent not only drives in lane but also handles unsafe conditions at an expert level, increasing robustness.We make our code publicly available.

翻译：本文提出了一种新颖的逆强化学习框架，利用基于扩散的自适应前瞻规划器（IRL-DAL）实现自动驾驶车辆的控制。训练始于对专家有限状态机（FSM）控制器的模仿学习，以提供稳定的初始化。环境反馈项与IRL判别器信号相结合，以对齐专家目标。随后，使用一种结合了扩散环境反馈与定向IRL奖励的混合奖励函数进行强化学习（RL）。一个作为安全监督器的条件扩散模型负责规划安全路径，确保车辆保持在车道内、规避障碍物并实现平稳行驶。接着，一个可学习的自适应掩码（LAM）用于提升感知能力，它能根据车速和附近危险动态调整视觉注意力。在基于FSM的模仿学习之后，策略通过近端策略优化（PPO）进行微调。训练在Webots仿真器中采用两阶段课程学习进行。该方法实现了96%的成功率，并将碰撞率降低至每千步0.05次，为安全导航设立了新的基准。应用所提出的方法后，智能体不仅能够实现车道内行驶，还能以专家水平处理不安全状况，从而显著增强了系统的鲁棒性。我们已公开相关代码。