MANTRA: Memory Augmented Networks for Multiple Trajectory Prediction

TL;DR

This paper introduces MANTRA, a memory-augmented neural network for multimodal trajectory prediction in autonomous driving, which leverages external memory to improve prediction accuracy and adapt to new data over time.

Contribution

The paper presents a novel memory-augmented neural network architecture that enhances multimodal trajectory prediction and allows continuous learning from new data in autonomous vehicle scenarios.

Findings

Achieves state-of-the-art results on three datasets.

Effectively incorporates scene knowledge via CNN.

Enables continuous improvement through external memory.

Abstract

Autonomous vehicles are expected to drive in complex scenarios with several independent non cooperating agents. Path planning for safely navigating in such environments can not just rely on perceiving present location and motion of other agents. It requires instead to predict such variables in a far enough future. In this paper we address the problem of multimodal trajectory prediction exploiting a Memory Augmented Neural Network. Our method learns past and future trajectory embeddings using recurrent neural networks and exploits an associative external memory to store and retrieve such embeddings. Trajectory prediction is then performed by decoding in-memory future encodings conditioned with the observed past. We incorporate scene knowledge in the decoding state by learning a CNN on top of semantic scene maps. Memory growth is limited by learning a writing controller based on the predictive capability of existing embeddings. We show that our method is able to natively perform multi-modal trajectory prediction obtaining state-of-the art results on three datasets. Moreover, thanks to the non-parametric nature of the memory module, we show how once trained our system can continuously improve by ingesting novel patterns.