MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction
This addresses the need for efficient and accurate probabilistic behavior prediction in autonomous driving, representing an incremental improvement over existing methods.
The paper tackles the problem of predicting multi-modal human behavior trajectories, particularly for autonomous driving, by introducing MultiPath, which uses a fixed set of anchor trajectories to efficiently output a Gaussian mixture distribution. The model achieves more accurate predictions and requires an order of magnitude fewer trajectories than sampling baselines.
Predicting human behavior is a difficult and crucial task required for motion planning. It is challenging in large part due to the highly uncertain and multi-modal set of possible outcomes in real-world domains such as autonomous driving. Beyond single MAP trajectory prediction, obtaining an accurate probability distribution of the future is an area of active interest. We present MultiPath, which leverages a fixed set of future state-sequence anchors that correspond to modes of the trajectory distribution. At inference, our model predicts a discrete distribution over the anchors and, for each anchor, regresses offsets from anchor waypoints along with uncertainties, yielding a Gaussian mixture at each time step. Our model is efficient, requiring only one forward inference pass to obtain multi-modal future distributions, and the output is parametric, allowing compact communication and analytical probabilistic queries. We show on several datasets that our model achieves more accurate predictions, and compared to sampling baselines, does so with an order of magnitude fewer trajectories.