Model Predictive Control for Crowd Navigation via Learning-Based Trajectory Prediction

TL;DR

This paper presents a deep learning-based trajectory predictor integrated with MPC for autonomous robot navigation in crowded environments, demonstrating improved safety and prediction accuracy over traditional models.

Contribution

It introduces a novel combination of Social-Implicit trajectory prediction with MPC, showing significant improvements in real-world pedestrian navigation tasks.

Findings

SI reduces prediction errors by up to 76% in low-density settings

SI-MPC enhances safety and motion smoothness in crowded scenes

Open-loop metrics may not fully reflect closed-loop navigation performance

Abstract

Safe navigation in pedestrian-rich environments remains a key challenge for autonomous robots. This work evaluates the integration of a deep learning-based Social-Implicit (SI) pedestrian trajectory predictor within a Model Predictive Control (MPC) framework on the physical Continental Corriere robot. Tested across varied pedestrian densities, the SI-MPC system is compared to a traditional Constant Velocity (CV) model in both open-loop prediction and closed-loop navigation. Results show that SI improves trajectory prediction - reducing errors by up to 76% in low-density settings - and enhances safety and motion smoothness in crowded scenes. Moreover, real-world deployment reveals discrepancies between open-loop metrics and closed-loop performance, as the SI model yields broader, more cautious predictions. These findings emphasize the importance of system-level evaluation and highlight the SI-MPC framework's promise for safer, more adaptive navigation in dynamic, human-populated environments.