Recursive Least Squares Based Refinement Network for the Rollout Trajectory Prediction Methods

TL;DR

This paper introduces a recursive least squares-based neural network module that enhances the accuracy and stability of rollout trajectory predictions in intelligent vehicle systems, addressing errors and adaptability issues.

Contribution

It presents a novel RLS estimation module integrated into deep neural networks, improving trajectory prediction robustness and long-term stability in rollout methods.

Findings

Improved prediction accuracy on NGSIM dataset

Enhanced long-term stability of trajectory predictions

Effective reduction of accumulative errors

Abstract

Trajectory prediction plays a pivotal role in the field of intelligent vehicles. It currently suffers from several challenges,e.g., accumulative error in rollout process and weak adaptability in various scenarios. This paper proposes a parametric-learning recursive least squares (RLS) estimation based on deep neural network for trajectory prediction. We design a flexible plug-in module which can be readily implanted into rollout approaches. Goal points are proposed to capture the long-term prediction stability from the global perspective. We carried experiments out on the NGSIM dataset. The promising results indicate that our method could improve rollout trajectory prediction methods effectively.