Application of Optimization Method for Identifying Human Operator Model Parameters

Abstract

The development of automation technologies in various fields of human life poses the problem of studying human-machine systems interaction and the influence of human on machine actions. The varied human behavior and the complexity of his central nervous system organization are a disability to his mathematical description. However, despite this, the human operator has efficient properties of trainability, adaptation to task variables, and optimization of system parameters. These properties allow researchers to introduce various assumptions in modeling and use them to identify the human model parameters. The purpose of this paper is to propose an approach to human operator model parameters identification by systems optimization methods. The paper is aimed to obtain a model of the human operator, which acts as a closed-loop feedback controller. The human operator model is taken in the form of the delayed lead-lag filter. Based on the assumption that the human operator optimizes system quality index, employing the optimization procedure to obtain/identify the human operator dynamics. The controlled plant model is assumed to be known and its parameters are fixed. The paper presents the results of the optimization-based identification procedure where the McRuer model in the compensatory control system was used. The proposed algorithm for calculation of the cost function is described mathematically and presented in pseudocode. Its reliability is demonstrated by comparing the numerical values of the human operator model parameters with the experimental results given in the literature.