研究:V2X 低频观测下学习型轨迹预测仍优于运动学模型
Learned versus Kinematic Trajectory Prediction under Reduced Observation Frequency
这篇论文回答了一个很实际的问题:V2X 只给你 1-10 Hz 的位置数据时,学习型预测模型还香不香。答案是到 2 Hz 都能扛住,1 Hz 就明显掉链子。做车端感知预测的工程师值得看看实验设计,假设预注册加多种子统计挺规范。
arXiv 论文 2611.23198 研究自动驾驶轨迹预测在低采样频率下的表现:V2X 通信中车辆位置通常以 1-10 Hz 到达,而模型多在固定频率数据上训练。作者在 Argoverse 2 数据集上把焦点智能体 5 秒历史从 10 Hz 降采样到 5、2、1 Hz,预测 6 秒未来,不使用地图和邻近智能体,比较针对各频率训练的 LSTM-RNN 与 10 Hz 训练后经线性插值处理稀疏历史的同一网络,测试集包含 24,984 个场景,每个 RNN 用 10 个种子训练并给出 bootstrap 置信区间。
结果显示,频率专属 RNN 的平均位移误差在 5 Hz 增加 3.1%、2 Hz 增加 10.1%、1 Hz 增加 20.3%。与假设相反,它相对更优运动学模型(CV/CA)的优势并未在直线运动上消失,在各频率下对直线和转弯均保持 24-30% 的优势;对行人而言 CV 反而更准确。插值方案的 10 Hz 模型在 2 Hz 及以上最多只比频率专属模型差 1.3%,但在 1 Hz 时落后 11.5%。每条假设都只获得部分支持,论文提示实际部署中插值在极端低频下有明显代价。
Learned versus Kinematic Trajectory Prediction under Reduced Observation Frequency
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 thresholds were fixed before training, the main RNNs were trained with ten seeds each, and results on 24,984 test scenarios have bootstrap confidence intervals over scenarios and seeds. The average displacement error of the frequency-specific RNN grew by 3.1% at 5 Hz, 10.1% at 2 Hz, and 20.3% at 1 Hz. Contrary to our hypothesis, its advantage over the better kinematic model did not vanish on straight motion and was 24-30% on both straight motion and turns at every frequency, while for pedestrians CV was more accurate. The 10 Hz model with interpolation was at most 1.3% less accurate than the frequency-specific model down to 2 Hz but 11.5% less accurate at 1 Hz. Each hypothesis was only partly supported.
本条情报归属 AITOP「论文」核心追踪栏目。涉及核心实体与技术议题包括:轨迹预测、Argoverse 2、LSTM、V2X,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.LG」(@Nukhbala Gadzhiev, Alexander Morochko),经 AITOP 多维评分管线结构化提炼。