[论文] Learned versus Kinematic Trajectory Prediction under Reduced Observati...
研究领域: ML 作者: Nukhbala Gadzhiev, Alexander Morochko 发布时间: 2026-11-17 arXiv: 2611.23198
论文概要
研究领域: ML 作者: Nukhbala Gadzhiev, Alexander Morochko 发布时间: 2026-11-17 arXiv: 2611.23198
中文摘要
自动驾驶轨迹预测模型通常在固定频率数据上训练,而 V2X 通信位置数据以 1-10 Hz 到达。我们研究仅依赖观察历史的学习预测器,在频率下降时是否保持对恒速(CV)和恒加速(CA)模型的优势。Argoverse 2 上,将 5 秒历史从 10 Hz 降至 5/2/1 Hz,预测 6 秒未来。频率特定 RNN 的平均位移误差在 5 Hz 时增 3.1%,2 Hz 时 10.1%,1 Hz 时 20.3%。与假设相反,其对运动学模型的优势在直线运动中没有消失,在所有频率下直线和转弯均保持 24-30%;对行人 CV 更准确。10 Hz 模型加插值在 2 Hz 以上精度损失≤1.3%,但 1 Hz 时低 11.5%。
原文摘要
Trajectory prediction models for automated driving are usually trained on data with a fixed sampling frequency, while positions received via vehicle-to-everything (V2X) communication arrive at 1-10 Hz. We study whether a learned predictor that uses only the observation history keeps its advantage over constant velocity (CV) and constant acceleration (CA) models when the history frequency drops. On the Argoverse 2 motion forecasting dataset, the 5 s history of the focal agent was downsampled from 10 Hz to 5, 2, and 1 Hz, and the 6 s future was predicted without maps or neighboring agents. An LSTM-based recurrent neural network (RNN) trained for each frequency was compared with the same network trained at 10 Hz and applied to sparse histories after linear interpolation. Hypotheses and thresh...
*自动采集于 2026-11-19*
#论文 #arXiv #ML #小凯