DynoSLAM: Dynamic SLAM with Generative Graph Neural Networks for Real-World Social Navigation

2026-05-04

AI Summary

This work addresses the performance degradation of traditional SLAM systems in dynamic environments, which typically rely on static-world assumptions. The authors propose DynoSLAM, a novel approach that tightly integrates generative graph neural networks with factor graph optimization. By constructing a stochastic world model of pedestrian motion and incorporating Monte Carlo rollouts, DynoSLAM explicitly captures the multimodal uncertainty inherent in human interactions through dynamic Mahalanobis distance factors. This formulation avoids deterministic heuristic predictions, thereby mitigating optimization failures caused by the “argmax problem” and enabling the computation of probabilistic safety margins for motion planning. Experimental results demonstrate that DynoSLAM achieves high-precision trajectory tracking in dense crowd simulations, significantly enhancing robotic navigation robustness and safety in highly dynamic scenes.

Abstract

Traditional Simultaneous Localization and Mapping (SLAM) algorithms rely heavily on the static environment assumption, which severely limits their applicability in real-world spaces populated by moving entities, such as pedestrians. In this work, we propose DynoSLAM, a tightly-coupled Dynamic GraphSLAM architecture that integrates socially-aware Graph Neural Networks (GNNs) directly into the factor graph optimization. Unlike conventional approaches that use rigid constant-velocity heuristics or deterministic single-agent neural priors, our framework formulates pedestrian motion forecasting as a stochastic World Model. By utilizing Monte Carlo rollouts from a trained GNN, we capture the multimodal epistemic uncertainty of human interactions and embed it into the SLAM graph via a dynamic Mahalanobis distance factor. We demonstrate through extensive simulated experiments that this stochastic formulation not only maintains highly accurate retrospective tracking but also prevents the optimization failures caused by the deterministic "argmax problem". Ultimately, extracting the empirical mean and covariance matrices of future pedestrian states provides a mathematically rigorous, probabilistic safety envelope for downstream local planners, enabling anticipatory and collision-free robot navigation in densely crowded environments.

Problem

• Dynamic SLAM
• social navigation
• moving entities
• static environment assumption
• pedestrian motion

Innovation

• Dynamic SLAM
• Graph Neural Networks
• Stochastic World Model
• Social Navigation
• Factor Graph Optimization