Accurate prediction of human or vehicle trajectories with good diversity that captures their stochastic nature is an essential task for many applications. However, many trajectory prediction models produce unreasonable trajectory samples that focus on improving diversity or accuracy while neglecting other key requirements, such as collision avoidance with the surrounding environment. In this work, we propose TrajDiffuse, a planning-based trajectory prediction method using a novel guided conditional diffusion model. We form the trajectory prediction problem as a denoising impaint task and design a map-based guidance term for the diffusion process. TrajDiffuse is able to generate trajectory predictions that match or exceed the accuracy and diversity of the SOTA, while adhering almost perfectly to environmental constraints. We demonstrate the utility of our model through experiments on the nuScenes and PFSD datasets and provide an extensive benchmark analysis against the SOTA methods.

翻译：准确预测具有良好多样性并能捕捉其随机性的人或车辆轨迹，是许多应用中的关键任务。然而，许多轨迹预测模型生成的轨迹样本不合理，它们侧重于提高多样性或准确性，却忽略了其他关键要求，例如与周围环境的碰撞避免。在这项工作中，我们提出了TrajDiffuse，一种基于规划的轨迹预测方法，它采用了一种新颖的引导条件扩散模型。我们将轨迹预测问题构建为一个去噪修复任务，并为扩散过程设计了一个基于地图的引导项。TrajDiffuse能够生成在准确性和多样性上匹配或超越当前最优（SOTA）水平的轨迹预测，同时几乎完美地遵守环境约束。我们通过在nuScenes和PFSD数据集上的实验，以及与SOTA方法的广泛基准分析，证明了我们模型的实用性。