Integrating kinematics and environment context into deep inverse reinforcement learning for predicting off-road vehicle trajectories

TL;DR

This paper introduces a novel two-stage neural network approach that integrates kinematic data and environmental context within an inverse reinforcement learning framework to improve off-road vehicle trajectory prediction.

Contribution

The work presents a new method combining environment features and kinematics in IRL for more accurate trajectory prediction, addressing previous separation of these factors.

Findings

Effective extraction of environmental and kinematic features from off-road data

Accurate prediction of multi-modal future trajectories at intersections

Different predictions based on vehicle speed at the same intersection

Abstract

Predicting the motion of a mobile agent from a third-person perspective is an important component for many robotics applications, such as autonomous navigation and tracking. With accurate motion prediction of other agents, robots can plan for more intelligent behaviors to achieve specified objectives, instead of acting in a purely reactive way. Previous work addresses motion prediction by either only filtering kinematics, or using hand-designed and learned representations of the environment. Instead of separating kinematic and environmental context, we propose a novel approach to integrate both into an inverse reinforcement learning (IRL) framework for trajectory prediction. Instead of exponentially increasing the state-space complexity with kinematics, we propose a two-stage neural network architecture that considers motion and environment together to recover the reward function. The first-stage network learns feature representations of the environment using low-level LiDAR statistics and the second-stage network combines those learned features with kinematics data. We collected over 30 km of off-road driving data and validated experimentally that our method can effectively extract useful environmental and kinematic features. We generate accurate predictions of the distribution of future trajectories of the vehicle, encoding complex behaviors such as multi-modal distributions at road intersections, and even show different predictions at the same intersection depending on the vehicle's speed.