From Goals, Waypoints & Paths To Long Term Human Trajectory Forecasting
This work tackles the problem of predicting human trajectories over significantly longer durations for applications like autonomous driving and robotics, where understanding future human movement is critical for safety and planning.
This paper addresses the challenge of long-term human trajectory forecasting by factorizing uncertainty into epistemic (long-term goals) and aleatoric (waypoints and paths) sources. The proposed Y-net model significantly improves state-of-the-art performance on both short prediction horizons (Stanford Drone & ETH/UCY datasets) and a novel long prediction horizon setting (up to one minute) on repurposed Stanford Drone & Intersection Drone datasets.
Human trajectory forecasting is an inherently multi-modal problem. Uncertainty in future trajectories stems from two sources: (a) sources that are known to the agent but unknown to the model, such as long term goals and (b)sources that are unknown to both the agent & the model, such as intent of other agents & irreducible randomness indecisions. We propose to factorize this uncertainty into its epistemic & aleatoric sources. We model the epistemic un-certainty through multimodality in long term goals and the aleatoric uncertainty through multimodality in waypoints& paths. To exemplify this dichotomy, we also propose a novel long term trajectory forecasting setting, with prediction horizons upto a minute, an order of magnitude longer than prior works. Finally, we presentY-net, a scene com-pliant trajectory forecasting network that exploits the pro-posed epistemic & aleatoric structure for diverse trajectory predictions across long prediction horizons.Y-net significantly improves previous state-of-the-art performance on both (a) The well studied short prediction horizon settings on the Stanford Drone & ETH/UCY datasets and (b) The proposed long prediction horizon setting on the re-purposed Stanford Drone & Intersection Drone datasets.