Shared Decision-Making Forward an Autonomous Navigation for Intelligent Vehicles

Shared control is a situation where humans and machines carry out a task in a cooperative way, and it is a popular approach to facilitate control and communication between humans and intelligent systems. Automated driving vehicles that share control with human drivers have been shown to acts better than fully automated systems and manual driving. However, there is little consensus in guidelines for the design and definition of shared control, some researchers assist humans using haptic feedback, and others use an optimization process and many other works. In this paper, a non-haptic shared control is applied to navigate an intelligent vehicle based on two inputs one by a human driver and the other obtained by an autonomous algorithm. Where, Non-Cooperative Game Theory is applied to formulate the final input of the vehicle in the form of linear velocity and steering wheel angle; where a joystick is used for manipulating human inputs and a computer vision technology "Visual Servoing" is implemented to command the autonomous driving inputs, after calculating the fusion coefficient value, which is determined by a function dependent on the admissibility of inputs and the similarity between them. The final navigation decision sent to the intelligent vehicle is the output of the fusion system. Noting that, it has the only right to command the driving inputs to the vehicle. Finally, simulation results are presented with the help of SCANeR Studio and MATLAB/SIMULINK Simulators.