Research map / Probabilistic and scientific modeling / Dynamical systems
Learning Control-Oriented Dynamical Structure from Data
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Learn SDC factorizations with the dynamics model and use SD-LQR | Control-affine nonlinear systems · PVTOL | RMS tracking error: N = 100; N_traj = 5 | Helped |
| Learn SDC factorizations with the dynamics model and use SD-LQR | Control-affine nonlinear systems · Spacecraft | RMS tracking error | Mixed |
| Learn SDC factorizations with the dynamics model and use SD-LQR | Control-affine nonlinear systems · PVTOL double loop-the-loop trajectory | tracking success | Helped |
| Learn SDC factorizations alongside the dynamics model, then use linearized LQR instead of SD-LQR (ablation) | SDC-regularized learned dynamics model · Spacecraft and PVTOL | RMS tracking error | Helped |
| Use the learned SDC factorizations directly in SD-LQR rather than only using the SDC-regularized model with linearized LQR (ablation) | SDC-regularized learned dynamics model · Trajectory data, N_traj in {1,2} | RMS tracking error: N_traj ∈ {1,2} | Mixed |
| Learn a controller and CCM jointly using sampled certificate-condition penalties | CCM learning · Spacecraft and PVTOL | RMS tracking error | Hurt |
| Learn a dynamics model for linearized LQR rather than learning SDC structure | Naive LQR learning · PVTOL double loop-the-loop trajectory | tracking success | Hurt |
| Increase labelled training data for naive learned linearized LQR | Naive LQR learning · Spacecraft and PVTOL | RMS tracking error | Helped |
| Use a learned nonlinear error-dynamics factorization instead of the oracle LQR's linear approximation | SD-LQR · PVTOL | RMS tracking error: N = 100; N_traj = 5 | Helped |
Closest papers by meaning
- Neural Time-Reversed Generalized Riccati Equation (AAAI 2024 · similarity 0.86)
- Performative Control for Linear Dynamical Systems (NeurIPS 2024 · similarity 0.86)
- Analytical Lyapunov Function Discovery: An RL-based Generative Approach (ICML 2025 · similarity 0.85)
- Two‑Stage Learning of Stabilizing Neural Controllers via Zubov Sampling and Iterative Domain Expansion (NeurIPS 2025 · similarity 0.85)
- Markovian Dynamics Enforcer: Feasibility Preserving Correction on Learned Dyanmics Manifolds (NeurIPS 2026 · similarity 0.84)
- A Memory Efficient Unified Algorithm for Online Learning of Linear Dynamical Systems (NeurIPS 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.