Real-Time Learning of Predictive Dynamic Obstacle Models for Robotic Motion Planning

TL;DR

This paper introduces a real-time, noise-robust method for predicting the motions of dynamic agents using a modified Hankel-DMD approach, enabling better motion forecasting for autonomous systems.

Contribution

It develops a novel online framework combining Hankel-DMD, SVHT, and structured low-rank constraints for real-time dynamic obstacle prediction.

Findings

Achieves stable variance-aware denoising in simulations and experiments.

Provides accurate short-horizon motion predictions for dynamic agents.

Enables integration into real-time robotic control systems.

Abstract

Autonomous systems often must predict the motions of nearby agents from partial and noisy data. This paper asks and answers the question: "can we learn, in real-time, a nonlinear predictive model of another agent's motions?" Our online framework denoises and forecasts such dynamics using a modified sliding-window Hankel Dynamic Mode Decomposition (Hankel-DMD). Partial noisy measurements are embedded into a Hankel matrix, while an associated Page matrix enables singular-value hard thresholding (SVHT) to estimate the effective rank. A Cadzow projection enforces structured low-rank consistency, yielding a denoised trajectory and local noise variance estimates. From this representation, a time-varying Hankel-DMD lifted linear predictor is constructed for multi-step forecasts. The residual analysis provides variance-tracking signals that can support downstream estimators and risk-aware planning. We validate the approach in simulation under Gaussian and heavy-tailed noise, and experimentally on a dynamic crane testbed. Results show that the method achieves stable variance-aware denoising and short-horizon prediction suitable for integration into real-time control frameworks.