用生成世界模型缓解自动驾驶模仿学习中的分布偏移问题
Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Models
- 在潜空间构建生成世界模型，让策略学会从错误中恢复
- 在CARLA和NVIDIA DRIVE Sim中均显著优于现有方法
- 适合关注自动驾驶鲁棒性与模仿学习的从业者
本文提出使用潜空间生成世界模型来解决自动驾驶模仿学习中的协变量偏移问题。世界模型是一种神经网络，可基于历史状态和动作预测智能体的下一状态。通过在训练中引入世界模型，驾驶策略无需大量数据即可有效缓解协变量偏移。在端到端训练过程中，策略通过与人类示范中观察到的状态对齐，学会从错误中恢复，从而在运行时能够应对训练分布外的扰动。此外，我们引入了一种基于Transformer的感知编码器，采用多视角交叉注意力机制和可学习场景查询。实验结果表明，在CARLA仿真器的闭环测试中显著优于现有最先进方法，并在CARLA和NVIDIA DRIVE Sim中均展现出对扰动的处理能力。
原文摘要 · Abstract (English)
We propose the use of latent space generative world models to address the covariate shift problem in autonomous driving. A world model is a neural network capable of predicting an agent's next state given past states and actions. By leveraging a world model during training, the driving policy effectively mitigates covariate shift without requiring an excessive amount of training data. During end-to-end training, our policy learns how to recover from errors by aligning with states observed in human demonstrations, so that at runtime it can recover from perturbations outside the training distribution. Additionally, we introduce a novel transformer-based perception encoder that employs multi-view cross-attention and a learned scene query. We present qualitative and quantitative results, demonstrating significant improvements upon prior state of the art in closed-loop testing in the CARLA simulator, as well as showing the ability to handle perturbations in both CARLA and NVIDIA's DRIVE Sim.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。