SWMLP: Shared Weight Multilayer Perceptron for Car Trajectory Speed Prediction using Road Topographical Features

TL;DR

This paper introduces SWMLP, a novel speed prediction model using road topographical features and shared weights in a multilayer perceptron, addressing data scarcity and regional variability in traffic data.

Contribution

The paper presents a new shared weight multilayer perceptron model that predicts vehicle speeds using topographical features without relying on extensive historical data.

Findings

Significant improvement over standard regression methods

Effective in regions with limited historical traffic data

Provides a new approach for traffic analysis using topographical features

Abstract

Although traffic is one of the massively collected data, it is often only available for specific regions. One concern is that, although there are studies that give good results for these data, the data from these regions may not be sufficiently representative to describe all the traffic patterns in the rest of the world. In quest of addressing this concern, we propose a speed prediction method that is independent of large historical speed data. To predict a vehicle's speed, we use the trajectory road topographical features to fit a Shared Weight Multilayer Perceptron learning model. Our results show significant improvement, both qualitative and quantitative, over standard regression analysis. Moreover, the proposed framework sheds new light on the way to design new approaches for traffic analysis.