Leveraging Smooth Attention Prior for Multi-Agent Trajectory Prediction
Leveraging Smooth Attention Prior for Multi-Agent Trajectory Prediction
复制标题
DOI：
10.48550/arxiv.2203.04421
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Zhangjie Cao;Erdem Biyik;G. Rosman;Dorsa Sadigh
中科院分区：
文献类型：
--
作者：
Zhangjie Cao;Erdem Biyik;G. Rosman;Dorsa Sadigh
多智能体交互对于预测其他智能体的行为和轨迹的建模是重要的。在某个时候，为了预测合理的未来轨迹，每个智能体只需要关注与一小群最相关的智能体的交互，而不是不必要地关注所有其他智能体。然而，现有的注意力建模工作忽略了人类在驾驶中的注意力不会迅速变化，并且可能会在时间步长上引入波动的注意力。在本文中，我们制定了一个注意力模型的基础上，多智能体的相互作用的总变化时间平滑先验，并提出了一个轨迹预测架构，利用这些出席的相互作用的知识。我们演示了如何总变异注意力先验沿着与新的序列预测损失项导致更平滑的注意力和更多的样本有效的学习多智能体轨迹预测，并通过比较它与最先进的方法在合成和自然驾驶数据显示其在预测精度方面的优势。我们在我们的网站11 https：//www.example.com上展示了我们的算法在INTERACTION数据集上的轨迹预测性能。sites.google.com/view/smoothness-attention
Multi-agent interactions are important to model for forecasting other agents' behaviors and trajectories. At a certain time, to forecast a reasonable future trajectory, each agent needs to pay attention to the interactions with only a small group of most relevant agents instead of unnecessarily paying attention to all the other agents. However, existing attention modeling works ignore that human attention in driving does not change rapidly, and may introduce fluctuating attention across time steps. In this paper, we formulate an attention model for multi-agent interactions based on a total variation temporal smoothness prior and propose a trajectory prediction architecture that leverages the knowledge of these attended interactions. We demonstrate how the total variation attention prior along with the new sequence prediction loss terms leads to smoother attention and more sample-efficient learning of multi-agent trajectory prediction, and show its advantages in terms of prediction accuracy by comparing it with the state-of-the-art approaches on both synthetic and naturalistic driving data. We demonstrate the performance of our algorithm for trajectory prediction on the INTERACTION dataset on our website11https://sites.google.com/view/smoothness-attention.