Tube-NeRF：基于管道引导数据增强与NeRF的MPC视觉运动策略高效模仿学习
Tube-NeRF: Efficient Imitation Learning of Visuomotor Policies from MPC using Tube-Guided Data Augmentation and NeRFs
AI总结：
本文提出Tube-NeRF方法，结合鲁棒MPC与NeRF数据增强，高效训练多旋翼飞行器视觉运动策略，实现80倍演示效率提升和50%训练时间缩减，并成功迁移至真实平台。
摘要：
模仿学习（IL）可以从资源密集型的模型预测控制器（MPC）中训练计算高效的感知运动策略，但通常需要大量样本，导致训练时间过长或鲁棒性有限。为解决这些问题，我们将IL与一种考虑过程和感知不确定性的鲁棒MPC变体相结合，并设计了一种数据增强（DA）策略，以实现基于视觉策略的高效学习。所提出的DA方法名为Tube-NeRF，利用神经辐射场（NeRFs）生成新颖的合成图像，并利用鲁棒MPC的性质（管道）来选择相关视图并高效计算相应的动作。我们将该方法针对多旋翼飞行器上的定位与轨迹跟踪任务进行了定制，通过学习一个视觉运动策略，该策略仅使用机载相机图像作为水平位置的唯一来源来生成控制动作。数值评估表明，与现有IL方法相比，演示效率提高了80倍，训练时间减少了50%。此外，我们的策略成功迁移到真实多旋翼飞行器上，在存在较大扰动的情况下实现了低跟踪误差，机载推理时间仅为1.5毫秒。
英文摘要：
Imitation learning (IL) can train computationally-efficient sensorimotor policies from a resource-intensive Model Predictive Controller (MPC), but it often requires many samples, leading to long training times or limited robustness. To address these issues, we combine IL with a variant of robust MPC that accounts for process and sensing uncertainties, and we design a data augmentation (DA) strategy that enables efficient learning of vision-based policies. The proposed DA method, named Tube-NeRF, leverages Neural Radiance Fields (NeRFs) to generate novel synthetic images, and uses properties of the robust MPC (the tube) to select relevant views and to efficiently compute the corresponding actions. We tailor our approach to the task of localization and trajectory tracking on a multirotor, by learning a visuomotor policy that generates control actions using images from the onboard camera as only source of horizontal position. Numerical evaluations show 80-fold increase in demonstration efficiency and a 50% reduction in training time over current IL methods. Additionally, our policies successfully transfer to a real multirotor, achieving low tracking errors despite large disturbances, with an onboard inference time of only 1.5 ms. Video: https://youtu.be/_W5z33ZK1m4
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
