Towards Explainable Robots: Developing Consensus Reaching Mechanisms for Co-Robots in Haptic Shared Control Paradigms

Human-automation teaming (HAT) is gaining importance in military and commercial applications with autonomous vehicles because it promises to improve performance, reduce the cost of operating and designing platforms, and increase adaptability to new situations. Given that both humans and automation systems are subject to misses, faults, or errors, to ensure the HAT performance in unpredictable conditions, it is critical to address the hand-off problem -- how to transition control between a human driver and automation system. Current solutions for control transfer in semi-automated ground vehicles face issues such as protracted transfer time, misinterpretations, or misappropriations of responsibility, and incomplete or inaccurate understandings of the vehicle and environment state. Transitions involving such issues are often "bumpy'' and implicated in safety compromises.
In this dissertation, we designed and tested an adaptive haptic shared control wherein a driver and an automation system are physically connected through a motorized steering wheel. We model the structure of the automation system like the structure of the human-driver, including a higher-level intent generator and lower-level impedance controller. In the first phase of the project, we developed a
nonlinear stochastic model predictive approach to determine how automation's impedance should be modulated in different interaction modes to enable the smooth and dynamic transition of control authority. Then, we tested our controller through a set of human-subject studies using a fixed-base driving simulator. Our findings showed that by adaptively modulating the impedance of the automation system, the control transfer time is reduced, and the performance of HAT is significantly improved.
In the second phase of this dissertation, we studied the principles of convention formation in a haptic shared control framework to narrow down the many possible strategies for resolving a conflict to those that a driver might be more gravitated. To this end, we proposed a modular platform to separate partner-specific conventions from task-dependent representations and use this platform to learn various forms of conventions between a human-driver and automation system. Using this platform, we will create a map from human-automation interaction outcomes to the space of conventions. This map will then be used to design an adaptable automation system. To design an adaptable automation system, we developed a reinforcement-learning model predictive controller wherein the characteristic of the model-predictive controller, including the weights of its cost function, is updated in different interaction modes using the learned convention map. Finally, we tested the proposed platform on the problem of intent negotiation between the driver and the automation system. The results demonstrated that the conflict between humans and automation could be further reduced using the convention-based approach.