Giving Santos™ Awareness: Cognitive Abilities
The VSR team is working closely with their partners in the industry and at the University of Iowa Hospitals and Clinics to create a brain for Santos™. The objective of this work is to enhance Santos's™ autonomous capabilities with human-like decision making. Santos™ will be completely independent, able to live in his virtual environment and have autonomous movements and decision making capabilities without any aid or external interference. Santos's™ decision- making process will be based on precedent information, learning processes, and future expectations. Toward this end, the VSR team is addressing two frontiers.
First, the VSR team will use neural networking as a main engine for Santos's™ decision making model. A neural network is a type of artificial intelligent that attempts to imitate the way a human brain works. Neural networking is particularly effective for predicting events when the networks have a large database, and they are currently used prominently in voice, image, and gesture recognition systems, as well as in industrial robotics and aerospace applications. For each assigned mission, Santos™ will use neural networking and view a mission as hierarchal structures with many layers of subtasks. With each subtask, Santos™ will use his optimization-based capabilities to select the optimal solution. For example, by feeding in a target goal, Santos™ can then use neural network's architecture to select the most appropriate motor commands to achieve the target.
The second part of our work will focus on adding emotion to Santos's™ character. This component will play an important role in the support and development of Santos's™ environment and social interaction. The VSR team proposes a computational cognitive model for Santos™ to acquire incrementally affective categories (e.g. sadness, anger, happiness) as part of his decision-making process. Toward this end, the VSR team is working closely with Dr. Antoine Bechara in the Neurology Department at the University of Iowa Hospital and Clinics, on implementing the University of Iowa somatic marker hypothesis into a computer based cognitive model which will be part of Santos decision making capabilities.
This cognitive model considers the role of different areas in the brain, such as the amygdale and the ventromedial prefrontal (VM) cortex, in the decision making process. Once positive and negative emotions are triggered, the model finds a summation of their net values. This summation is not algebraic. Rather, it depends on the laws of natural selection, i.e., stronger responses gain selective advantage over weaker ones, so that the strongest response dominates, and the weakest disappears.
There are several challenging extensions to be tackled here, that VSR team is working on, and the contributors to this special issue include neurologist, psychologists, engineers, and computer scientists.
