Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs

TL;DR

This paper introduces a novel spectral clustering and graph-LSTM based method for urban traffic forecasting, predicting both trajectories and behaviors of road-agents with significantly improved accuracy and long-term prediction performance.

Contribution

It combines spectral graph analysis with deep learning to enhance trajectory and behavior prediction accuracy in traffic scenarios, introducing a spectral regularization technique and theoretical error bounds.

Findings

Reduces average prediction error by ~75%

Achieves 91.2% accuracy in behavior prediction

Improves long-term prediction by up to 70%

Abstract

We present a novel approach for traffic forecasting in urban traffic scenarios using a combination of spectral graph analysis and deep learning. We predict both the low-level information (future trajectories) as well as the high-level information (road-agent behavior) from the extracted trajectory of each road-agent. Our formulation represents the proximity between the road agents using a weighted dynamic geometric graph (DGG). We use a two-stream graph-LSTM network to perform traffic forecasting using these weighted DGGs. The first stream predicts the spatial coordinates of road-agents, while the second stream predicts whether a road-agent is going to exhibit overspeeding, underspeeding, or neutral behavior by modeling spatial interactions between road-agents. Additionally, we propose a new regularization algorithm based on spectral clustering to reduce the error margin in long-term prediction (3-5 seconds) and improve the accuracy of the predicted trajectories. Moreover, we prove a theoretical upper bound on the regularized prediction error. We evaluate our approach on the Argoverse, Lyft, Apolloscape, and NGSIM datasets and highlight the benefits over prior trajectory prediction methods. In practice, our approach reduces the average prediction error by approximately 75% over prior algorithms and achieves a weighted average accuracy of 91.2% for behavior prediction. Additionally, our spectral regularization improves long-term prediction by up to 70%.