JRDB-Traj: A Dataset and Benchmark for Trajectory Forecasting in Crowds

TL;DR

This paper introduces JRDB-Traj, a comprehensive dataset and benchmark for end-to-end trajectory forecasting in crowds, emphasizing real-world scenarios with raw sensory data and a new evaluation metric for practical applications.

Contribution

The paper presents a novel dataset extending JRDB for trajectory forecasting, enabling end-to-end evaluation with raw sensory inputs and introduces a new metric for real-world scenario assessment.

Findings

Dataset includes agents, scene images, point clouds from robot perspective

Supports evaluation of models with imperfect tracking modules

Provides a benchmark for real-world trajectory forecasting

Abstract

Predicting future trajectories is critical in autonomous navigation, especially in preventing accidents involving humans, where a predictive agent's ability to anticipate in advance is of utmost importance. Trajectory forecasting models, employed in fields such as robotics, autonomous vehicles, and navigation, face challenges in real-world scenarios, often due to the isolation of model components. To address this, we introduce a novel dataset for end-to-end trajectory forecasting, facilitating the evaluation of models in scenarios involving less-than-ideal preceding modules such as tracking. This dataset, an extension of the JRDB dataset, provides comprehensive data, including the locations of all agents, scene images, and point clouds, all from the robot's perspective. The objective is to predict the future positions of agents relative to the robot using raw sensory input data. It bridges the gap between isolated models and practical applications, promoting a deeper understanding of navigation dynamics. Additionally, we introduce a novel metric for assessing trajectory forecasting models in real-world scenarios where ground-truth identities are inaccessible, addressing issues related to undetected or over-detected agents. Researchers are encouraged to use our benchmark for model evaluation and benchmarking.