Data-Driven Modeling and Correction of Vehicle Dynamics
Nguyen Ly, Caroline Tatsuoka, Jai Nagaraj, Jacob Levy, Fernando Palafox, David Fridovich-Keil, Hannah Lu
TL;DR
The paper tackles learning and correction of non-autonomous vehicle dynamics under model-form uncertainty by locally parameterizing time-varying inputs to yield sequences of autonomous sub-systems. It introduces DRIPS, a data-efficient framework that builds linear surrogates in lifted spaces with interpolation across parameter space, and combines it with Flow Map Learning (FML) for stronger nonlinearities, augmented by a transfer-learning-based correction mechanism that requires only scarce high-fidelity data. The authors validate the approach on unicycle and planar bicycle variants with and without slip, demonstrating robust predictive accuracy and effective model correction under data scarcity. The work provides a cohesive methodology to fuse physics-based vehicle models with data-driven corrections, with implications for reliable trajectory prediction and control in realistic, time-varying settings and potentially for multi-vehicle, multi-agent systems.
Abstract
We develop a data-driven framework for learning and correcting non-autonomous vehicle dynamics. Physics-based vehicle models are often simplified for tractability and therefore exhibit inherent model-form uncertainty, motivating the need for data-driven correction. Moreover, non-autonomous dynamics are governed by time-dependent control inputs, which pose challenges in learning predictive models directly from temporal snapshot data. To address these, we reformulate the vehicle dynamics via a local parameterization of the time-dependent inputs, yielding a modified system composed of a sequence of local parametric dynamical systems. We approximate these parametric systems using two complementary approaches. First, we employ the DRIPS (dimension reduction and interpolation in parameter space) methodology to construct efficient linear surrogate models, equipped with lifted observable spaces and manifold-based operator interpolation. This enables data-efficient learning of vehicle models whose dynamics admit accurate linear representations in the lifted spaces. Second, for more strongly nonlinear systems, we employ FML (Flow Map Learning), a deep neural network approach that approximates the parametric evolution map without requiring special treatment of nonlinearities. We further extend FML with a transfer-learning-based model correction procedure, enabling the correction of misspecified prior models using only a sparse set of high-fidelity or experimental measurements, without assuming a prescribed form for the correction term. Through a suite of numerical experiments on unicycle, simplified bicycle, and slip-based bicycle models, we demonstrate that DRIPS offers robust and highly data-efficient learning of non-autonomous vehicle dynamics, while FML provides expressive nonlinear modeling and effective correction of model-form errors under severe data scarcity.