Evaluating Pedestrian Trajectory Prediction Methods with Respect to Autonomous Driving

TL;DR

The paper benchmarks open-source pedestrian trajectory prediction methods on ETH/UCY with a focus on single-trajectory outputs, evaluating accuracy via ADE and FDE, sensitivity to motion-history length, and runtime scalability. It finds that while the constant velocity baseline remains a strong practical option for many dynamic scenes, graph-based models like Trajectron++ and Social-Implicit achieve the best Best-of-N accuracy, with Trajectron++ excelling in single-sample predictions. Reducing the observed history generally degrades performance, though some models gain from longer histories, indicating varying reliance on temporal features. The study provides actionable guidance for autonomous driving applications, suggesting hybrid architectures and explicit intention/semantic cues to handle state changes and static scenarios, and underscores the need for traffic-oriented datasets to better capture real-world driving conditions.

Abstract

In this paper, we assess the state of the art in pedestrian trajectory prediction within the context of generating single trajectories, a critical aspect aligning with the requirements in autonomous systems. The evaluation is conducted on the widely-used ETH/UCY dataset where the Average Displacement Error (ADE) and the Final Displacement Error (FDE) are reported. Alongside this, we perform an ablation study to investigate the impact of the observed motion history on prediction performance. To evaluate the scalability of each approach when confronted with varying amounts of agents, the inference time of each model is measured. Following a quantitative analysis, the resulting predictions are compared in a qualitative manner, giving insight into the strengths and weaknesses of current approaches. The results demonstrate that although a constant velocity model (CVM) provides a good approximation of the overall dynamics in the majority of cases, additional features need to be incorporated to reflect common pedestrian behavior observed. Therefore, this study presents a data-driven analysis with the intent to guide the future development of pedestrian trajectory prediction algorithms.

Figures (7)

• Figure 1: The three features commonly employed for pedestrian trajectory prediction: the motion history of each agent, spatial information, and interactions.
• Figure 2: ADE (blue) and FDE (orange) across various models using one and 20 sampled trajectories
• Figure 3: ADE (blue) and FDE (orange) across various models for $N = {1,2,8}$ observations
• Figure 4: Runtime of the pedestrian trajectory prediction models represented by their median inference time over all scenarios and their variability as interquartile range (IQR) between the 90th and 10th percentiles over the Average Displacement Error for $K=1$ and $N=8$.
• Figure 5: Selected samples from the ETH scenario, showcasing four commonly observed pedestrian motion patterns alongside the predictions of each investigated model.