Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces the Markovian Dynamics Enforcer (MaDE), a post-hoc operator designed to correct neural trajectory predictions that violate physical constraints, even when controls are unobserved. MaDE maps proposed state transitions onto a learned feasible dynamics manifold by inferring controls, recomputing states via a physics-informed completion model with a learned residual, and then iteratively correcting these controls via gradient-based inequality reduction. The method achieves near-zero dynamics residuals on fully specified systems and significantly reduces residuals on underspecified systems, outperforming raw predictors on real-world vehicle trajectory data.
Neural predictors can be made dynamically feasible without ground-truth controls by a frozen, post-hoc operator that enforces learned dynamics manifolds via iterative control correction.
Neural trajectory predictors can reach low prediction error while violating dynamics, actuator limits, or state constraints, especially when controls are unobserved and dynamics are partially specified. We introduce the Markovian Dynamics Enforcer (MaDE), a time-invariant post-hoc operator mapping state-transition proposals onto a learned feasible dynamics manifold, trained on feasible states without ground-truth controls. For each transition it infers a control and recomputes the state through a completion model of known physics plus a learned residual. It then corrects that control by gradient-based inequality reduction, so inequality satisfaction is best-effort within an iteration budget. Since every correction iterate re-enters the completion model, the returned state is dynamically consistent by construction relative to that model and the supplied previous-state anchor. MaDE drives dynamics residuals to essentially zero on fully specified simulated systems, and on an underspecified system leaves a smaller true-dynamics residual than the baselines. Designed to attach to arbitrary predictors, the frozen operator is evaluated downstream of recurrent, structured state-space, and transformer predictors. On recorded vehicle trajectories the one-step residual against a kinematic bicycle model is 0.0071 to 0.0072 for MaDE and 0.1703 to 0.1714 for raw predictors. MaDE raises average displacement error by a factor of 1.57 to 1.83.