Polar Collision Grids: Effective Interaction Modelling for Pedestrian Trajectory Prediction in Shared Space Using Collision Checks

TL;DR

This paper introduces a novel polar collision grid map for better modeling of pedestrian interactions, improving trajectory prediction accuracy in shared spaces by considering collision risk, approach direction, and velocity.

Contribution

It proposes a heuristic-based agent selection method using collision risk, incorporating approach direction and velocity into a new polar collision grid for enhanced prediction.

Findings

Predicted trajectories are closer to ground truth than baseline methods.

The approach effectively models pedestrian-vehicle and pedestrian-pedestrian interactions.

Improved accuracy demonstrated on the HBS dataset.

Abstract

Predicting pedestrians' trajectories is a crucial capability for autonomous vehicles' safe navigation, especially in spaces shared with pedestrians. Pedestrian motion in shared spaces is influenced by both the presence of vehicles and other pedestrians. Therefore, effectively modelling both pedestrian-pedestrian and pedestrian-vehicle interactions can increase the accuracy of the pedestrian trajectory prediction models. Despite the huge literature on ways to encode the effect of interacting agents on a pedestrian's predicted trajectory using deep-learning models, limited effort has been put into the effective selection of interacting agents. In the majority of cases, the interaction features used are mainly based on relative distances while paying less attention to the effect of the velocity and approaching direction in the interaction formulation. In this paper, we propose a heuristic-based process of selecting the interacting agents based on collision risk calculation. Focusing on interactions of potentially colliding agents with a target pedestrian, we propose the use of time-to-collision and the approach direction angle of two agents for encoding the interaction effect. This is done by introducing a novel polar collision grid map. Our results have shown predicted trajectories closer to the ground truth compared to existing methods (used as a baseline) on the HBS dataset.