Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From GEV to ResLogit: Spatially Correlated Discrete Choice Models for Pedestrian Movement Prediction

Rulla Al-Haideri, Bilal Farooq

TL;DR

The results suggest that in dense, high frequency spatial choice sets, learning based residual corrections can capture proximity induced correlation more effectively than analyst specified GEV nesting structures, while maintaining interpretability.

Abstract

High frequency pedestrian motion forecasting when interacting with autonomous vehicles (AVs) can be enhanced through the use of behavioural frameworks, such as discrete choice models, that can explicitly account for correlation among similar movement alternatives. We formulate the pedestrian next step choice as a spatial discrete choice defined by a grid of speed adjustment and heading change. Using naturalistic pedestrian-AV encounters from nuScenes and Argoverse 2 (1 sec decision interval), we estimate a multinomial logit baseline and four spatial generalized extreme value (GEV) specifications (SCL, GSCL, SCNL, and GSCNL). We then compare them to a residual neural network logit (ResLogit) model that learns cross alternative effects while retaining an interpretable linear utility component. Across the evaluated data, spatial GEV structures yield only marginal improvements over multinomial logit, whereas ResLogit achieves a substantially better fit and produces behaviourally coherent errors concentrated among neighbouring grid cells. The results suggest that in dense, high frequency spatial choice sets, learning based residual corrections can capture proximity induced correlation more effectively than analyst specified GEV nesting structures, while maintaining interpretability.

From GEV to ResLogit: Spatially Correlated Discrete Choice Models for Pedestrian Movement Prediction

TL;DR

The results suggest that in dense, high frequency spatial choice sets, learning based residual corrections can capture proximity induced correlation more effectively than analyst specified GEV nesting structures, while maintaining interpretability.

Abstract

High frequency pedestrian motion forecasting when interacting with autonomous vehicles (AVs) can be enhanced through the use of behavioural frameworks, such as discrete choice models, that can explicitly account for correlation among similar movement alternatives. We formulate the pedestrian next step choice as a spatial discrete choice defined by a grid of speed adjustment and heading change. Using naturalistic pedestrian-AV encounters from nuScenes and Argoverse 2 (1 sec decision interval), we estimate a multinomial logit baseline and four spatial generalized extreme value (GEV) specifications (SCL, GSCL, SCNL, and GSCNL). We then compare them to a residual neural network logit (ResLogit) model that learns cross alternative effects while retaining an interpretable linear utility component. Across the evaluated data, spatial GEV structures yield only marginal improvements over multinomial logit, whereas ResLogit achieves a substantially better fit and produces behaviourally coherent errors concentrated among neighbouring grid cells. The results suggest that in dense, high frequency spatial choice sets, learning based residual corrections can capture proximity induced correlation more effectively than analyst specified GEV nesting structures, while maintaining interpretability.
Paper Structure (15 sections, 19 equations, 4 figures, 6 tables)

This paper contains 15 sections, 19 equations, 4 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Pedestrian motion prediction
  4. Structured action spaces and discretization in motion forecasting
  5. Correlation across alternatives in discrete choice
  6. Discrete choice models for pedestrian motion and correlated alternatives
  7. Research gap
  8. Methods
  9. Dataset
  10. Model Structure
  11. Models Specification
  12. Results and Discussions
  13. Confusion matrices and classification metrics
  14. Interpretation of linear utility coefficients
  15. Conclusions

Figures (4)

  • Figure 1: Pedestrian spatial choice set grid.
  • Figure 2: Confusion matrices for the spatial GEV models using maximum probability (MaxP) predictions.
  • Figure 3: ResLogit confusion matrices on original and synthetic data.
  • Figure 4: Original versus synthetic distributions for ResLogit variables. The ddist and ddir plots are pooled across the nine alternatives.