Less is More: Contextual Sampling for Nonlinear Data-Enabled Predictive Control

2025-03-31

AI Summary

Nonlinear data-driven predictive control (DeePC) suffers from high computational overhead, hindering real-time deployment in robotic systems—particularly for motion planning and trajectory tracking—where accuracy and efficiency are fundamentally at odds. To address this, we propose a dynamic context-aware sampling strategy: instead of static or statistical data selection, our method leverages online similarity metrics and Hankel matrix modeling to dynamically identify the most relevant nonlinear sub-trajectories at each control step, enabling workload-adaptive dataset compression. This significantly reduces computational load without compromising prediction fidelity: in autonomous vehicle motion planning, it achieves a 53.2% reduction in tracking error versus leverage-score sampling; using only 100 trajectories (out of 491), peak computation time drops by 87.2% while maintaining comparable tracking accuracy. To our knowledge, this is the first integration of temporal-aware dynamic sampling into the DeePC framework, offering an efficient and practical solution for real-time nonlinear robotic control.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Data-enabled Predictive Control (DeePC) is a powerful data-driven approach for predictive control without requiring an explicit system model. However, its high computational cost limits its applicability to real-time robotic systems. For robotic applications such as motion planning and trajectory tracking, real-time control is crucial. Nonlinear DeePC either relies on large datasets or learning the nonlinearities to ensure predictive accuracy, leading to high computational complexity. This work introduces contextual sampling, a novel data selection strategy to handle nonlinearities for DeePC by dynamically selecting the most relevant data at each time step. By reducing the dataset size while preserving prediction accuracy, our method improves computational efficiency, of DeePC for real-time robotic applications. We validate our approach for autonomous vehicle motion planning. For a dataset size of 100 sub-trajectories, Contextual sampling DeePC reduces tracking error by 53.2 % compared to Leverage Score sampling. Additionally, Contextual sampling reduces max computation time by 87.2 % compared to using the full dataset of 491 sub-trajectories while achieving comparable tracking performance. These results highlight the potential of Contextual sampling to enable real-time, data-driven control for robotic systems.

Problem

• Reducing computational cost in Data-enabled Predictive Control
• Improving real-time performance for robotic applications
• Enhancing accuracy with dynamic data selection strategy

Innovation

• Dynamic data selection for nonlinear DeePC
• Reduces dataset size, maintains accuracy
• Enables real-time robotic control efficiently