Cognitive TransFuser: Semantics-guided Transformer-based Sensor Fusion for Improved Waypoint Prediction

TL;DR

Cognitive TransFuser is a multi-task sensor fusion network that leverages semantics-guided features from RGB and LIDAR data to improve waypoint prediction for autonomous driving, achieving real-time performance in simulation.

Contribution

The paper introduces a novel semantics-guided transformer-based sensor fusion approach that incorporates auxiliary tasks for enhanced waypoint prediction in autonomous driving.

Findings

Significant performance improvement over baseline in waypoint prediction.

Achieves up to 44.2 FPS real-time inference in CARLA simulator.

Effective multi-task fusion of semantic and auxiliary features enhances safety.

Abstract

Sensor fusion approaches for intelligent self-driving agents remain key to driving scene understanding given visual global contexts acquired from input sensors. Specifically, for the local waypoint prediction task, single-modality networks are still limited by strong dependency on the sensitivity of the input sensor, and thus recent works therefore promote the use of multiple sensors in fusion in feature level in practice. While it is well known that multiple data modalities encourage mutual contextual exchange, it requires global 3D scene understanding in real-time with minimal computation upon deployment to practical driving scenarios, thereby placing greater significance on the training strategy given a limited number of practically usable sensors. In this light, we exploit carefully selected auxiliary tasks that are highly correlated with the target task of interest (e.g., traffic light recognition and semantic segmentation) by fusing auxiliary task features and also using auxiliary heads for waypoint prediction based on imitation learning. Our RGB-LIDAR-based multi-task feature fusion network, coined Cognitive TransFuser, augments and exceeds the baseline network by a significant margin for safer and more complete road navigation in the CARLA simulator. We validate the proposed network on the Town05 Short and Town05 Long Benchmark through extensive experiments, achieving up to 44.2 FPS real-time inference time.