Predicting the future motion of traffic agents is vital for self-driving vehicles to ensure their safe operation. We introduce RedMotion, a transformer model for motion prediction that incorporates two types of redundancy reduction. The first type of redundancy reduction is induced by an internal transformer decoder and reduces a variable-sized set of road environment tokens, such as road graphs with agent data, to a fixed-sized embedding. The second type of redundancy reduction is a self-supervised learning objective and applies the redundancy reduction principle to embeddings generated from augmented views of road environments. Our experiments reveal that our representation learning approach can outperform PreTraM, Traj-MAE, and GraphDINO in a semi-supervised setting. Our RedMotion model achieves results that are competitive with those of Scene Transformer or MTR++. We provide an open source implementation that is accessible via GitHub (https://github.com/kit-mrt/red-motion) and Colab (https://colab.research.google.com/drive/1Q-Z9VdiqvfPfctNG8oqzPcgm0lP3y1il).

翻译：预测交通参与者的未来运动对自动驾驶车辆的安全运行至关重要。我们提出RedMotion，一种用于运动预测的Transformer模型，该模型引入了两种冗余约简机制。第一种冗余约简由内部Transformer解码器触发，将可变大小的道路环境令牌集合（例如包含智能体数据的道路图）约简为固定大小的嵌入表示。第二种冗余约简是一种自监督学习目标，将冗余约简原理应用于从道路环境增强视图生成的嵌入。实验表明，在半监督场景下，我们的表示学习方法可超越PreTraM、Traj-MAE和GraphDINO。RedMotion模型取得了与Scene Transformer或MTR++相竞争的结果。我们提供了开源实现，可通过GitHub（https://github.com/kit-mrt/red-motion）和Colab（https://colab.research.google.com/drive/1Q-Z9VdiqvfPfctNG8oqzPcgm0lP3y1il）访问。