This Small Business Innovation Research Phase I project focuses on optimizing the design of a robot for children with Autism Spectrum Disorders (ASD). Research has shown that children with ASD interact more easily with inanimate devices, such as robots, than with humans. Despite this promising research, there are currently no consumer robots on the market specifically designed to interact with ASD children. The goal of this research is to investigate what set of features are optimal for a robot designed to be an effective and affordable therapy tool for children with ASD. Initial designs will be generated based on existing research and previously conducted user tests. These designs will be used to create prototypes which will then be used to conduct user studies with ASD children to determine which designs elicit the greatest engagement, measured by proximity to robot, number of interactions (physical & eye contact), duration of interactions and number of vocalizations. The expected results are a validated design of an affordable consumer-ready robot that addresses the specific needs of children with ASD.
The broader impact/commercial potential of this project involve both the direct impact of providing a beneficial new therapy tool for children with ASD as well as providing a platform from which additional research can be launched. The CDC currently estimates that 1 in 88 children in the United States will be diagnosed with an Autism Spectrum Disorder and ASDs are reported to occur in all racial, ethnic, and socioeconomic groups. Early intervention can make a significant difference in the lives of these children and their families. Tools that allow children to practice and reinforce skills learned in formal therapy sessions are strongly desired by parents. The goal of this project is to develop a validated design for a robot to help children with ASD practice social referencing skills. By designing both for effectiveness and affordability, it will be the first autism robot targeted towards the home therapy market. Additionally, the results of this project will also be beneficial to researchers studying the use of robots as autism therapy tools. This robot will serve as an affordable alternative to expensive research robots, most of which were not designed specifically to interact with special needs children.
The Centers for Disease Control and Prevention currently estimates that 1 in 88 children born in the United States will be diagnosed with an Autism Spectrum Disorder (ASD). Children with autism struggle with a variety of social interaction skills including communication, eye contact, emotion recognition and emotion regulation. For the past several years Interbots LLC has been developing "Popchilla," a robotic therapy system that combines mobile apps, interactive robots, and a data tracking service to help children with autism practice important life skills through interactive play. Recently, touchscreen-based tablet devices have been garnering a lot of attention in the autism community for their ability to captivate the attention of children with autism and facilitate new forms of communication. In addition, research has shown that children with autism have an easier time interacting with mechanical devices, such as robots, than humans. By leveraging these technologies, the Popchilla system provides a fun and nonthreatening therapeutic platform for children with autism to practice social interaction and daily life skills. The overarching goal of this Phase I research project was to optimize the design of the Popchilla robot for children with autism. The technical goals included determining the optimal design of Popchilla’s ears, eyes, mouth and "skin" for maximizing engagement and therapeutic value, reducing motor noise, improving the head motion of the robot, and developing a new design of the robot based on the results of this research. To determine the optimal design of Popchilla’s ears, eyes, mouth and "skin," four user studies were conducted in partnership with the Quality of Life Technology Center at Carnegie Mellon University. An online emotional expression study was performed to determine which of Popchilla’s features are the strongest cues for emotional expressions, and the minimal feature set needed to correctly identify emotional expressions. Thirty-four participants with an autism diagnosis ages 3-17 completed the survey. Results revealed that the mouth was the most important feature for conveying emotion, followed by the eyes and ears, and only one expressive feature is required for emotion recognition. Several designs of Popchilla’s eyes, ears, mouth, color and texture were tested in combination to determine which combinations were most engaging to children with autism. Testing was conducted in-person. Participants were nineteen children diagnosed with autism or an autism-related disorder divided into two populations: ages 4-9 with more severe symptoms and ages 11-15 with less severe symptoms. The younger participants preferred non-moving ears and non color-changing eyes, while the older participants favored moving ears and color-changing eyes. The younger participants were more engaged by animated eyes and mouth on a screen, while the older participants found physical eyes and a physical mouth more compelling. No one fur color was strongly preferred among either population, though pink, green and purple were the most popular choices among the younger participants while blue, yellow and red were the most popular among the older participants. Similarly, no one texture was preferred among either population, though the younger participants showed a preference for non-fur textures while the older participants favored fur of varying lengths. Internal testing revealed that encasing motors in foam reduces the average motor noise volume by 63% and the peak volume by 79%. Research also revealed that the range and quality of head motion could be improved by separating the head fur from the body fur and eliminating the head up/down motion and replacing it with a waist lean forward/backward motion. Overall, results of the user studies were quite varied. Engagement with the various eyes, ears and mouth designs was evenly split among the participants and there was no clear color or texture preference. This is not surprising given the diversity in behavioral and emotional response among children with autism. Rather, it emphasizes that as much as possible, products designed for populations with ASD should support multi-modal features and customizability. Based on the results of these studies as well as other factors such as manufacturing costs, manufacturing complexity and difficulty of animation and content generation, a new (proprietary) design of the Popchilla robot was developed. As a result of funding from the National Science Foundation and our valuable scientific, technical and community partnerships, Interbots aims to become the leader in autism robotics, providing parents and therapists with software, robotic tools and curricula to support the social, emotional and cognitive development of children with autism. By developing the first robotic therapy system for children with autism, Interbots hopes to improve the lives of families struggling with this debilitating disorder. This Phase I research and our future Phase II project are invaluable in helping us achieve our commercial and humanitarian goals.
