Multi-Predictor Fusion: Combining Learning-based and Rule-based Trajectory Predictors
This work addresses the need for reliable trajectory prediction in autonomous vehicles, particularly in interactive traffic scenarios, though it appears incremental as it builds on existing predictors.
The paper tackles the problem of improving trajectory prediction for autonomous vehicles by combining learning-based and rule-based predictors, resulting in a method that outperforms standalone predictors on various metrics and delivers more consistent performance.
Trajectory prediction modules are key enablers for safe and efficient planning of autonomous vehicles (AVs), particularly in highly interactive traffic scenarios. Recently, learning-based trajectory predictors have experienced considerable success in providing state-of-the-art performance due to their ability to learn multimodal behaviors of other agents from data. In this paper, we present an algorithm called multi-predictor fusion (MPF) that augments the performance of learning-based predictors by imbuing them with motion planners that are tasked with satisfying logic-based rules. MPF probabilistically combines learning- and rule-based predictors by mixing trajectories from both standalone predictors in accordance with a belief distribution that reflects the online performance of each predictor. In our results, we show that MPF outperforms the two standalone predictors on various metrics and delivers the most consistent performance.