Robotics Agent Coacher for CP motor Function (RAC CP Fun)

A gap of several decades exists between the latest technological achievements in humanoid robots, machine learning algorithms, and artificial intelligence, and the applications used by professional caregivers, especially in the field of pediatric rehabilitation. Robotics Agent Coacher for Cerebral Palsy motor Function (RAC CP Fun) is an attempt to implement advanced humanoid robotic technology, principles of Socially Assistive Robotics (SAR), and a motor learning approach in rehabilitation of movement disorders of central origin. Application of the recent advances in these areas offers an opportunity for a leap forward in rehabilitation practice. RAC CP Fun targets children with Cerebral Palsy (CP), and is designed to improve their motor function and activities of daily living. In terms of International Classification of Functioning, Disability, and Health (ICF) accepted by the World Health Organization, the novel algorithms of RAC CP Fun aim to facilitate children's activity and participation through pleasurable game-like human-robot interaction. At the same time, these algorithms utilize the motor learning approach, which is based on principles of repetition, stages of learning, appropriate feedback, random practice, and enriched environments. In this paper we introduce the experimental architecture, scenarios, and architectural design of RAC CP Fun, and findings from pilot experiments. Implementation of socially assistive robotics in neuromuscular rehabilitation may also stimulate the development of new treatment interventions for a wide variety of diseases and disorders through effective physiotherapy practice.