Reliably Learn to Trim Multiparametric Quadratic Programs via Constraint Removal

TL;DR

This paper introduces a method to efficiently reduce the complexity of multiparametric quadratic programs by learning to remove redundant constraints, significantly speeding up solutions in control applications.

Contribution

It proposes novel learning-based methods to reliably trim mp-QPs through constraint removal, applicable online and offline, especially for model predictive control systems.

Findings

Redundant constraints can be effectively identified and removed.

The number of constraints in MPC mp-QPs can be reduced to zero over time.

Simulation results show significant efficiency improvements.

Abstract

In a wide range of applications, we are required to rapidly solve a sequence of convex multiparametric quadratic programs (mp-QPs) on resource-limited hardwares. This is a nontrivial task and has been an active topic for decades in control and optimization communities. Observe that the main computational cost of existing solution algorithms lies in addressing many linear inequality constraints, though their majority are redundant and removing them will not change the optimal solution. This work learns from the results of previously solved mp-QP(s), based on which we propose novel methods to reliably trim (unsolved) mp-QPs via constraint removal, and the trimmed mp-QPs can be much cheaper to solve. Then, we extend to trim mp-QPs of model predictive control (MPC) whose parameter vectors are sampled from linear systems. Importantly, both online and offline solved mp-QPs can be utilized to adaptively trim mp-QPs in the closed-loop system. We show that the number of linear inequalities in the trimmed mp-QP of MPC decreases to zero in a finite timestep, which also can be reduced by increasing offline computation. Finally, simulations are performed to demonstrate the efficiency of our trimming method in removing redundant constraints.