Certainty Equivalent Perception-Based Control

In order to certify performance and safety, feedback control requires precise characterization of sensor errors. In this paper, we provide guarantees on such feedback systems when sensors are characterized by solving a supervised learning problem. We show a uniform error bound on nonparametric kernel regression under a dynamically-achievable dense sampling scheme. This allows for a finite-time convergence rate on the sub-optimality of using the regressor in closed-loop for waypoint tracking. We demonstrate our results in simulation with simplified unmanned aerial vehicle and autonomous driving examples.

[1]  Nikolai Matni,et al.  System Level Synthesis , 2019, Annu. Rev. Control..

[2]  Andreas Krause,et al.  Safe Model-based Reinforcement Learning with Stability Guarantees , 2017, NIPS.

[3]  Alexey Dosovitskiy,et al.  End-to-End Driving Via Conditional Imitation Learning , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[4]  Nevena Lazic,et al.  Model-Free Linear Quadratic Control via Reduction to Expert Prediction , 2018, AISTATS.

[5]  Benjamin Recht,et al.  The Gap Between Model-Based and Model-Free Methods on the Linear Quadratic Regulator: An Asymptotic Viewpoint , 2018, COLT.

[6]  Vijay Kumar,et al.  Aggressive Flight With Suspended Payloads Using Vision-Based Control , 2018, IEEE Robotics and Automation Letters.

[7]  Csaba Szepesvári,et al.  Online Least Squares Estimation with Self-Normalized Processes: An Application to Bandit Problems , 2011, ArXiv.

[8]  Victor Chernozhukov,et al.  Exact and Robust Conformal Inference Methods for Predictive Machine Learning With Dependent Data , 2018, COLT.

[9]  Byron Boots,et al.  Deep Forward and Inverse Perceptual Models for Tracking and Prediction , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[10]  Koen Tiels,et al.  Identification of Nonlinear Block-Oriented Systems starting from Linear Approximations: A Survey , 2016, ArXiv.

[11]  Samira Kamoun,et al.  Combined Parameter and State Estimation Algorithms for Multivariable Nonlinear Systems Using MIMO Wiener Models , 2016 .

[12]  Nikolai Matni,et al.  Regret Bounds for Robust Adaptive Control of the Linear Quadratic Regulator , 2018, NeurIPS.

[13]  Nikolai Matni,et al.  Safely Learning to Control the Constrained Linear Quadratic Regulator , 2018, 2019 American Control Conference (ACC).

[14]  Nikolai Matni,et al.  Scalable system level synthesis for virtually localizable systems , 2017, 2017 IEEE 56th Annual Conference on Decision and Control (CDC).

[15]  Soon-Jo Chung,et al.  Robust Regression for Safe Exploration in Control , 2019, L4DC.

[16]  Kim Peter Wabersich,et al.  Linear Model Predictive Safety Certification for Learning-Based Control , 2018, 2018 IEEE Conference on Decision and Control (CDC).

[17]  Michael I. Jordan,et al.  Active Learning for Nonlinear System Identification with Guarantees , 2020, J. Mach. Learn. Res..

[18]  Wlodzimierz Greblicki,et al.  Nonparametric approach to Wiener system identification , 1997 .

[19]  Larry Wasserman,et al.  Distribution‐free prediction bands for non‐parametric regression , 2014 .

[20]  Z. Hasiewicz.  Identification of a linear system observed through zero-memory non-linearity , 1987 .

[21]  Hannelore Liero,et al.  Strong uniform consistency of nonparametric regression function estimates , 1989 .

[22]  Dennis S. Bernstein,et al.  Subspace identification for nonlinear systems that are linear in unmeasured states , 2001, Proceedings of the 40th IEEE Conference on Decision and Control (Cat. No.01CH37228).

[23]  J. Pearson,et al.  l^{1} -optimal feedback controllers for MIMO discrete-time systems , 1987 .

[24]  Nikolai Matni,et al.  Structured state space realizations for SLS distributed controllers , 2017, 2017 55th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[25]  Csaba Szepesvári,et al.  Regret Bounds for the Adaptive Control of Linear Quadratic Systems , 2011, COLT.

[26]  Gábor Orosz,et al.  End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks , 2019, AAAI.

[27]  Nikolai Matni,et al.  Robust Guarantees for Perception-Based Control , 2019, L4DC.

[28]  Soon-Jo Chung,et al.  Chance-Constrained Trajectory Optimization for Safe Exploration and Learning of Nonlinear Systems , 2020, IEEE Robotics Autom. Lett..

[29]  Claire Tomlin,et al.  Eyes-Closed Safety Kernels: Safety for Autonomous Systems Under Loss of Observability , 2020, Robotics: Science and Systems.

[30]  Max Simchowitz,et al.  Improper Learning for Non-Stochastic Control , 2020, COLT.

[31]  T. Wigren.  Convergence analysis of recursive identification algorithms based on the nonlinear Wiener model , 1994, IEEE Trans. Autom. Control..

[32]  B. Hansen.  UNIFORM CONVERGENCE RATES FOR KERNEL ESTIMATION WITH DEPENDENT DATA , 2008, Econometric Theory.

[33]  Akshay Krishnamurthy,et al.  Learning the Linear Quadratic Regulator from Nonlinear Observations , 2020, NeurIPS.

[34]  Nikolai Matni,et al.  On the Sample Complexity of the Linear Quadratic Regulator , 2017, Foundations of Computational Mathematics.

[35]  Nikolai Matni,et al.  A System-Level Approach to Controller Synthesis , 2016, IEEE Transactions on Automatic Control.

[36]  Max Simchowitz,et al.  Learning Linear Dynamical Systems with Semi-Parametric Least Squares , 2019, COLT.

[37]  L. Devroye.  The uniform convergence of the nadaraya‐watson regression function estimate , 1978 .

[38]  Max Simchowitz,et al.  Naive Exploration is Optimal for Online LQR , 2020, ICML.

[39]  Akshay Krishnamurthy,et al.  Kinematic State Abstraction and Provably Efficient Rich-Observation Reinforcement Learning , 2019, ICML.

[40]  Nikolai Matni,et al.  Finite-Data Performance Guarantees for the Output-Feedback Control of an Unknown System , 2018, 2018 IEEE Conference on Decision and Control (CDC).

[41]  J. M. M. Montiel,et al.  ORB-SLAM: A Versatile and Accurate Monocular SLAM System , 2015, IEEE Transactions on Robotics.

[42]  G. Bennett.  Probability Inequalities for the Sum of Independent Random Variables , 1962 .

[43]  Germán Ros,et al.  CARLA: An Open Urban Driving Simulator , 2017, CoRL.

[44]  Benjamin Recht,et al.  Certainty Equivalence is Efficient for Linear Quadratic Control , 2019, NeurIPS.

[45]  E. Candès,et al.  The limits of distribution-free conditional predictive inference , 2019, Information and Inference: A Journal of the IMA.

[46]  Johan Schoukens,et al.  Wiener system identification with generalized orthonormal basis functions , 2014, Autom..

[47]  Aaron D. Ames,et al.  Episodic Learning with Control Lyapunov Functions for Uncertain Robotic Systems* , 2019, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).