Human-Intelligent System Shared Control Strategy with Conflict Resolution

Abstract

Shared control strategies play a very important role in the Human-Intelligent system collaboration thus bridging the gap between them. It gives the scope to bring out the best in human and intelligent system (IntelSys). Different successful shared control strategies developed in the past, limit the scope of collaboration or put restrictions on the independent human and IntelSys behavior. In this paper, we have presented a new generic shared control framework by introducing the idea of a separate fusion system with the objective to blend human-IntelSys individual control inputs. Only fusion system has the direct control of the plant system. The fusion system is built with a new approach, which is based on quantifying and analyzing the difference in the individual control inputs which is termed as ’Conflict’. Shared control is achieved by resolving this conflict which is expressed in the form of a linear quadratic cost function. The shared controller is designed as an adaptive linear quadratic regulator. The generic nature of this framework covers various types of use cases with respect to the admissibility or correctness of the individual control inputs. The shared control strategy is validated on the simulated inverted pendulum system in the simulated environment using Simulink software.