DiPA: Probabilistic Multi-Modal Interactive Prediction for Autonomous Driving

TL;DR

DiPA is a novel probabilistic multi-modal prediction method for autonomous driving that effectively captures diverse future behaviors, improves accuracy, and supports planning in interactive scenarios.

Contribution

DiPA introduces a Gaussian-Mixture-Model based approach with a novel training regime to better encode full distributions and balance probabilistic accuracy with behavior diversity.

Findings

Achieves state-of-the-art results on INTERACTION and NGSIM datasets.

Outperforms baseline in both probabilistic and closest-mode evaluations.

Effectively supports planning in interactive driving scenarios.

Abstract

Accurate prediction is important for operating an autonomous vehicle in interactive scenarios. Prediction must be fast, to support multiple requests from a planner exploring a range of possible futures. The generated predictions must accurately represent the probabilities of predicted trajectories, while also capturing different modes of behaviour (such as turning left vs continuing straight at a junction). To this end, we present DiPA, an interactive predictor that addresses these challenging requirements. Previous interactive prediction methods use an encoding of k-mode-samples, which under-represents the full distribution. Other methods optimise closest-mode evaluations, which test whether one of the predictions is similar to the ground-truth, but allow additional unlikely predictions to occur, over-representing unlikely predictions. DiPA addresses these limitations by using a Gaussian-Mixture-Model to encode the full distribution, and optimising predictions using both probabilistic and closest-mode measures. These objectives respectively optimise probabilistic accuracy and the ability to capture distinct behaviours, and there is a challenging trade-off between them. We are able to solve both together using a novel training regime. DiPA achieves new state-of-the-art performance on the INTERACTION and NGSIM datasets, and improves over the baseline (MFP) when both closest-mode and probabilistic evaluations are used. This demonstrates effective prediction for supporting a planner on interactive scenarios.