The Trajectron: Probabilistic Multi-Agent Trajectory Modeling With Dynamic Spatiotemporal Graphs
About
Developing safe human-robot interaction systems is a necessary step towards the widespread integration of autonomous agents in society. A key component of such systems is the ability to reason about the many potential futures (e.g. trajectories) of other agents in the scene. Towards this end, we present the Trajectron, a graph-structured model that predicts many potential future trajectories of multiple agents simultaneously in both highly dynamic and multimodal scenarios (i.e. where the number of agents in the scene is time-varying and there are many possible highly-distinct futures for each agent). It combines tools from recurrent sequence modeling and variational deep generative modeling to produce a distribution of future trajectories for each agent in a scene. We demonstrate the performance of our model on several datasets, obtaining state-of-the-art results on standard trajectory prediction metrics as well as introducing a new metric for comparing models that output distributions.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Trajectory Prediction | ETH UCY (test) | ADE0.22 | 65 | |
| Trajectory Prediction | ZARA1 v1.0 (test) | ADE0.17 | 58 | |
| Trajectory Prediction | ETH-UCY | -- | 57 | |
| Trajectory Prediction | Hotel ETH-UCY (test) | ADE0.12 | 48 | |
| Trajectory Prediction | ZARA2 v1.0 (test) | ADE0.12 | 36 | |
| Trajectory Prediction | ETH v1.0 (test) | ADE0.59 | 33 | |
| Pedestrian trajectory prediction | ETH (test) | -- | 29 | |
| Trajectory Forecasting | TrajNet++ real world Type III (test) | ADE0.6 | 19 | |
| Trajectory Forecasting | Hotel ETH scene (test) | ADE0.35 | 12 | |
| Trajectory Forecasting | Univ UCY scene (test) | ADE0.54 | 12 |