Safe Learning Predictive Control for Ego-World Robotic Systems

2026-07-24

AI Summary

This work addresses the challenge of safe autonomous navigation in shared environments, where the behaviors of other robots are governed by unknown and dynamic policies. To this end, the authors propose SOWL-MPC, a novel approach that integrates sparse variational Gaussian processes (SVGP) with online variational conditioning (OVC) within an ego-world framework. This integration enables, for the first time, real-time online learning of the posterior distribution over unknown policies and propagates policy uncertainty into model predictive control (MPC) via approximate moment propagation. The method substantially enhances collision avoidance safety and navigation robustness. Extensive Monte Carlo simulations on the ROS 2 platform demonstrate its real-time performance and safety, and the approach has been successfully deployed on a physical indoor robotic system.

Abstract

Safe autonomous navigation in shared environments requires the ability to anticipate and react to the latent behaviors of surrounding robots. In this paper, we propose SOWL-MPC, a safe learning-based predictive control strategy for a novel scenario, which we name ego-world robotic framework. In this setting, the control policy of the world robot is unknown and the ego exploits data to learn it and perform safe maneuvers. The proposed architecture combines an online learning mechanism based on Sparse Variational Gaussian Processes (SVGPs) with a receding-horizon control scheme. Relying solely on noisy state measurements, our approach infers a posterior distribution over the latent world policy, which is updated on streaming data via Online Variational Conditioning (OVC). The learned policy is propagated through the nonlinear world dynamics using an approximate moment propagation scheme, and fed to an uncertainty-aware Model Predictive Control (MPC), thus enabling safe maneuvering of the ego robot. The real-time feasibility and safety guarantees of SOWL-MPC are demonstrated through extensive Monte Carlo virtual experiments in ROS 2, and validated on real-world robotic hardware in an indoor arena.

Problem

• safe autonomous navigation
• latent behavior prediction
• ego-world interaction
• unknown control policy
• robotic safety

Innovation

• Safe Learning-Based Control
• Sparse Variational Gaussian Processes
• Online Variational Conditioning
• Uncertainty-Aware MPC
• Ego-World Robotic Systems