Trajectory Prediction using Equivariant Continuous Convolution
Trajectory Prediction using Equivariant Continuous Convolution
复制标题
DOI：
--
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
R. Walters;Jinxi Li;Rose Yu
中科院分区：
文献类型：
--
作者：
R. Walters;Jinxi Li;Rose Yu
轨迹预测是许多人工智能应用的关键部分，例如自动驾驶汽车的安全操作。然而，目前的方法容易做出不一致和物理上不切实际的预测。我们利用流体动力学的见解，考虑内部对称的轨迹，以克服这一限制。我们提出了一种新的模型，等变连续卷积（ECCO）改进的轨迹预测。ECCO使用旋转等变连续卷积来嵌入系统的对称性。在两个真实世界的车辆和行人轨迹数据集上，ECCO以更少的参数获得了具有竞争力的精度。它也是更有效的样本，从任何方向的几个数据点自动概括。最后，ECCO通过等方差改进了泛化，从而产生了物理上更一致的预测。我们的方法为提高深度学习模型的信任度和透明度提供了一个新的视角。
Trajectory prediction is a critical part of many AI applications, for example, the safe operation of autonomous vehicles. However, current methods are prone to making inconsistent and physically unrealistic predictions. We leverage insights from fluid dynamics to overcome this limitation by considering internal symmetry in trajectories. We propose a novel model, Equivariant Continous COnvolution (ECCO) for improved trajectory prediction. ECCO uses rotationally-equivariant continuous convolutions to embed the symmetries of the system. On two real-world vehicle and pedestrian trajectory datasets, ECCO attains competitive accuracy with significantly fewer parameters. It is also more sample efficient, generalizing automatically from few data points in any orientation. Lastly, ECCO improves generalization with equivariance, resulting in more physically consistent predictions. Our method provides a fresh perspective towards increasing trust and transparency in deep learning models.