Technical Report for Real-Time Certified Probabilistic Pedestrian Forecasting
This addresses the need for safe navigation in human-centric environments for autonomous systems, though it appears incremental as it builds on existing forecasting methods with specific improvements.
The paper tackles the problem of real-time probabilistic forecasting for pedestrians in autonomous systems by proposing a novel algorithm based on weighted sums of ordinary differential equations learned from historical trajectories, achieving considerably higher prediction quality than existing state-of-the-art approaches over long time horizons.
The success of autonomous systems will depend upon their ability to safely navigate human-centric environments. This motivates the need for a real-time, probabilistic forecasting algorithm for pedestrians, cyclists, and other agents since these predictions will form a necessary step in assessing the risk of any action. This paper presents a novel approach to probabilistic forecasting for pedestrians based on weighted sums of ordinary differential equations that are learned from historical trajectory information within a fixed scene. The resulting algorithm is embarrassingly parallel and is able to work at real-time speeds using a naive Python implementation. The quality of predicted locations of agents generated by the proposed algorithm is validated on a variety of examples and considerably higher than existing state of the art approaches over long time horizons.