Multiple-input and multiple-output force control for automobile component road simulation using model predictive control with proportional–integral–derivative

Abstract

When developing an entire vehicle system, testing the structure of the vehicle or each component as a module or individually is necessary to determine the reliability and ensure the endurance of the entire vehicle. Various tests have been conducted to check the durability of the parts. However, the most important part is the verification of the fatigue limit of the load vibration from the road surface when the vehicle is being driven. Verification can be achieved by experimenting while driving on a real road with a prototype vehicle best suited to the actual conditions. However, issues such as problems in time, space, and environmental constraints, inconsistency in driving characteristics of the test driver, and continuous monitoring exist. For testing the load vibration of the road surface in automobile parts in the laboratory, hydraulic servo actuators are used because they provide vibrational loads in multiple directions by configuring them in multiple axes rather than a single axis. In this article, a multiple-input multiple-output model predictive control–proportional–integral–derivative hybrid controller is proposed as the method for optimal control of a multi-axis hydraulic servo actuator used in a random road signal reproduction experiment. Its performance is compared with the simple proportional–integral–derivative controller. A method for obtaining an efficient black box multiple-input multiple-output system model using LabVIEW in a laboratory in the field is also introduced, and the effectiveness of the model predictive control–proportional–integral–derivative hybrid controller is shown by reproducing the actual road load.