Beyond Constant Parameters: Hyper Prediction Models and HyperMPC

2025-08-08

AI Summary

Traditional gradient-based Model Predictive Control (MPC) suffers from reduced long-horizon prediction accuracy due to oversimplified dynamical models that fail to capture unmodeled dynamics. To address this, we propose HyperPM—a hyper-predictive model that pioneers the integration of time-varying, learnable parameters into differential dynamical modeling. Specifically, a neural network explicitly learns the temporal evolution of these parameters over the MPC prediction horizon, projecting unmodeled dynamics onto a lightweight, time-varying parameter space. This approach preserves computational efficiency while significantly enhancing modeling flexibility and long-term prediction fidelity. We validate HyperMPC on complex real-world systems, including the F1TENTH autonomous racing platform. Experimental results demonstrate that HyperMPC substantially reduces long-horizon prediction error compared to state-of-the-art methods and achieves superior closed-loop control performance.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Auto ML and Hyperparameter Tuning
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Model Predictive Control (MPC) is among the most widely adopted and reliable methods for robot control, relying critically on an accurate dynamics model. However, existing dynamics models used in the gradient-based MPC are limited by computational complexity and state representation. To address this limitation, we propose the Hyper Prediction Model (HyperPM) - a novel approach in which we project the unmodeled dynamics onto a time-dependent dynamics model. This time-dependency is captured through time-varying model parameters, whose evolution over the MPC prediction horizon is learned using a neural network. Such formulation preserves the computational efficiency and robustness of the base model while equipping it with the capacity to anticipate previously unmodeled phenomena. We evaluated the proposed approach on several challenging systems, including real-world F1TENTH autonomous racing, and demonstrated that it significantly reduces long-horizon prediction errors. Moreover, when integrated within the MPC framework (HyperMPC), our method consistently outperforms existing state-of-the-art techniques.

Problem

• Improving robot control accuracy with dynamic models
• Reducing computational complexity in MPC prediction
• Enhancing long-horizon prediction for unmodeled phenomena

Innovation

• Hyper Prediction Model with time-varying parameters
• Neural network learns parameter evolution
• HyperMPC integrates HyperPM for improved control