Abstract:Optimal control problems for nonsmooth dynamical systems governed by differential variational inequalities (DVI) are called optimal control problems with equilibrium constraints (OCPEC). It provides a general formalism for nonsmooth optimal control. However, solving OCPEC using the direct method (i.e., first-discretize-then-optimize) is challenging owing to the lack of correct sensitivity and constraint regularity. This study uses the direct method to solve OCPEC and overcomes the numerical difficulties from two aspects: In the discretization step, we propose a class of novel approaches using gap functions to smooth the DVI, where gap functions are initially proposed for solving variational inequalities. The generated smoothing approximations of discretized OCPEC are called gap-constraint-based reformulations, which have a concise and semismoothly differentiable constraint system; In the optimization step, we propose an efficient dynamical system approach to solve the discretized OCPEC, where a sequence of gap-constraint-based reformulations is solved approximately. This dynamical system approach involves a semismooth Newton flow and achieves local exponential convergence under standard assumptions. The benchmark test shows that the proposed method is computationally tractable and achieves fast local convergence.

What problem does this paper attempt to address?

This paper attempts to solve the difficulties in the numerical solution of the Optimal Control Problem with Equilibrium Constraints (OCPEC). Specifically, the paper mainly focuses on the optimal control problems of non - smooth dynamic systems described by Differential Variational Inequalities (DVI). The direct method (i.e., discretize first and then optimize) faces two major challenges when solving such problems: 1. **Problems in the discretization step**: - The time - stepping method can only achieve first - order accuracy. - The numerical sensitivity is incorrect, that is, the gradient information of the discretized OCPEC does not match that of the continuous - time OCPEC, resulting in many artificial local minima in the discretized OCPEC. 2. **Problems in the optimization step**: - The discretized OCPEC is a difficult Nonlinear Programming (NLP) problem, known as the Mathematical Programming with Equilibrium Constraints problem (MPEC), which violates almost all Constraint Qualification conditions (CQs), making it difficult to apply the standard NLP theory. To solve these problems, the paper proposes the following two improvements: 1. **Smoothing DVI using gap functions**: - A new class of methods based on gap functions is proposed to smooth DVI. These methods were originally proposed for solving Variational Inequalities (VI). By introducing gap functions, DVI can be converted into a set of relaxed inequalities, thus generating a smoother approximation of the discretized OCPEC that is easier to solve. - Compared with the smoothing method based on the KKT conditions, this method does not need to introduce Lagrange multipliers, so the generated smoothing approximation has a smaller scale. 2. **Proposing an efficient dynamic system method**: - A dynamic system method based on semi - smooth Newton flow is proposed to solve the discretized OCPEC. This method can achieve local exponential convergence under standard assumptions and is solved through a series of reconstruction problems based on gap constraints. - This method utilizes the well - established dynamic system theory and can balance the efficiency and accuracy of computing intermediate iterations while ensuring feasibility and convergence. Through these improvements, the paper aims to provide an effective and reliable numerical method for solving the optimal control problem with equilibrium constraints, especially when dealing with non - smooth dynamic systems.

Dynamical System Approach for Optimal Control Problems with Equilibrium Constraints Using Gap-Constraint-Based Reformulation

A Successive Gap Constraint Linearization Method for Optimal Control Problems with Equilibrium Constraints

An Improved Smoothing Technique-Based Control Vector Parameterization Method for Optimal Control Problems with Inequality Path Constraints

A Quasi-Sequential Algorithm for PDE-constrained Optimization Based on Space–time Orthogonal Collocation on Finite Elements

Direct Approaches for PDE-constrained Dynamic Optimization Based on Space–time Orthogonal Collocation on Finite Elements

POD-based Reduced Order Optimal Control of Parabolic PDE Systems Via Diffusivity-Interior-boundary Actuation

Optimal Bounded Control of Stochastically Excited MDOF Nonlinear Viscoelastic Systems

A Novel Penalty Approach for Nonlinear Dynamic Optimization Problems with Inequality Path Constraints

A quasi-sequential approach to large-scale dynamic optimization problems

A non-interior-point continuation method for the optimal control problem with equilibrium constraints

Computation of Optimal Control Problems with Terminal Constraint via Modified Evolution Partial Differential Equation

Generic Homotopic Smoothing for Low-Thrust Optimal Control Problems With Power Constraints

On the Approximate Optimal Feedback Control for Dynamical Systems Modeled by Ordinary Differential Equations

A symplectic indirect approach for a class of nonlinear optimal control problems of differential‐algebraic systems

A New Smoothing Technique for Bang-Bang Optimal Control Problems

Discrete‐Time Optimal Control of State‐Constrained Nonlinear Systems Using Approximate Dynamic Programming

Dynamical System Approach for Time-Varying Constrained Convex Optimization Problems

Cooptimizing Safety and Performance with a Control-Constrained Formulation

Transform the Non-linear Programming Problem to the Initial-value Problem to Solve

Iterative Chebyshev approximation method for optimal control problems