AReN: assured ReLU NN architecture for model predictive control of LTI systems

In this paper, we consider the problem of automatically designing a Rectified Linear Unit (ReLU) Neural Network (NN) architecture that is sufficient to implement the optimal Model Predictive Control (MPC) strategy for an LTI system with quadratic cost. Specifically, we propose AReN, an algorithm to generate Assured ReLU Architectures. AReN takes as input an LTI system with quadratic cost specification, and outputs a ReLU NN architecture with the assurance that there exist network weights that exactly implement the associated MPC controller. AReN thus offers new insight into the design of ReLU NN architectures for the control of LTI systems: instead of training a heuristically chosen NN architecture on data - or iterating over many architectures until a suitable one is found - AReN can suggest an adequate NN architecture before training begins. While several previous works were inspired by the fact that ReLU NN controllers and optimal MPC controllers are both Continuous, Piecewise-Linear (CPWL) functions, exploiting this similarity to design NN architectures with correctness guarantees has remained elusive. AReN achieves this using two novel features. First, we reinterpret a recent result about the implementation of CPWL functions via ReLU NNs to show that a CPWL function may be implemented by a ReLU architecture that is determined by the number of distinct affine regions in the function. Second, we show that we can efficiently over-approximate the number of affine regions in the optimal MPC controller without solving the MPC problem exactly. Together, these results connect the MPC problem to a ReLU NN implementation without explicitly solving the MPC: the result is a NN architecture that has the assurance that it can implement the MPC controller. We show through numerical results the effectiveness of AReN in designing an NN architecture.

[1]  Frank Allgöwer,et al.  Learning an Approximate Model Predictive Controller With Guarantees , 2018, IEEE Control Systems Letters.

[2]  Fabian Pedregosa,et al.  Hyperparameter optimization with approximate gradient , 2016, ICML.

[3]  Byron Boots,et al.  Differentiable MPC for End-to-end Planning and Control , 2018, NeurIPS.

[4]  L. Chua,et al.  A generalized canonical piecewise-linear representation , 1990 .

[5]  Evangelos Theodorou,et al.  MPC-Inspired Neural Network Policies for Sequential Decision Making , 2018, ArXiv.

[6]  R. Stanley.  An Introduction to Hyperplane Arrangements , 2007 .

[7]  Yoshua Bengio,et al.  Random Search for Hyper-Parameter Optimization , 2012, J. Mach. Learn. Res..

[8]  Christian Tjandraatmadja,et al.  Bounding and Counting Linear Regions of Deep Neural Networks , 2017, ICML.

[9]  Riccardo Scattolini,et al.  Neural Network Implementation of Nonlinear Receding-Horizon Control , 1999, Neural Computing & Applications.

[10]  Hannu T. Toivonen,et al.  A neural network model predictive controller , 2006 .

[11]  Alberto Bemporad,et al.  The explicit linear quadratic regulator for constrained systems , 2003, Autom..

[12]  J. M. Tarela,et al.  Region configurations for realizability of lattice Piecewise-Linear models , 1999 .

[13]  Sriram Sankaranarayanan,et al.  Trajectory Tracking Control for Robotic Vehicles Using Counterexample Guided Training of Neural Networks , 2019, ICAPS.

[14]  Shimon Whiteson,et al.  Fast Efficient Hyperparameter Tuning for Policy Gradients , 2019, NeurIPS.

[15]  Ramesh Raskar,et al.  Designing Neural Network Architectures using Reinforcement Learning , 2016, ICLR.

[16]  Isabelle Guyon,et al.  Taking Human out of Learning Applications: A Survey on Automated Machine Learning , 2018, 1810.13306.

[17]  Shuning Wang,et al.  Generalization of hinging hyperplanes , 2005, IEEE Transactions on Information Theory.

[18]  Benjamin Karg,et al.  Efficient Representation and Approximation of Model Predictive Control Laws via Deep Learning , 2018, IEEE Transactions on Cybernetics.

[19]  Raman Arora,et al.  Understanding Deep Neural Networks with Rectified Linear Units , 2016, Electron. Colloquium Comput. Complex..

[20]  Vijay Kumar,et al.  Approximating Explicit Model Predictive Control Using Constrained Neural Networks , 2018, 2018 Annual American Control Conference (ACC).

[21]  Xin Zhang,et al.  End to End Learning for Self-Driving Cars , 2016, ArXiv.

[22]  Eduardo F. Camacho,et al.  Mobile robot navigation in a partially structured static environment, using neural predictive control , 1996 .

[23]  Paulo Tabuada,et al.  SMC: Satisfiability Modulo Convex Programming , 2018, Proceedings of the IEEE.

[24]  Paulo Tabuada,et al.  SMC: Satisfiability Modulo Convex Optimization , 2017, HSCC.

[25]  D. Luenberger.  Optimization by Vector Space Methods , 1968 .