Robot Sign Language Tutor for Children with Autism

Futurice, client: Satakunta hospital district
Article about the project (in Finnish)

Shortlisted for IXDA Interaction Design Awards 2019 - Disrupting category (winner selection ongoing)

Role: design lead

This project was the topic of my master's thesis, which was evaluated with the grade 5 (with 5 being the best, 1 the worst and 0 a failing grade). Ville Kyrki, who is an associate professor at Aalto University and heads the Intelligent Robotics research group, was the supervisor of this thesis.

The robot InMoov was named Momo for the purpose of this implementation. The original design of the InMoov has been made by Gael Langevin.


The robot InMoov had been built at Futurice previously. Satakunta hospital district was interested in using the robot to teach assistive sign language to children with autism. Children with autism experience communication problems, with 50 % never learning to speak. Those who do not speak well can use sign language to help them communicate.

Robots have previously been used in communication therapy with autistic children. This is due to the facts that autistic children have been shown to be more engaged in therapy if it includes technological or robotic components.

The robot InMoov had an initial design, which had to be modified for the purpose of teaching sign language to children with autism. Two experts from Satakunta hospital district, a speech therapist and a neuropsychologist, gave their expert opinions on design decisions.

The InMoov robot with a tablet implanted into its chest (one of the modifications developed for this project).

In order to design the robot, I developed a framework for the design process of social robots. To my knowledge, no such frameworks exist prior to this. This is also the first instance of a robot being used to teach sign language to children with autism.


The robot was modified according to 5 defined design guidelines. To test the resulting design, experiments with 10 autistic children were organized, in collaboration with Satakunta hospital district.

A blog post written about the project by my co-worker, Teemu Turunen, is available here. I will write a more complete blog post about the design process at a later date, after the thesis has been completed.

The robot interacting with the child (on the right), with the speech therapist present (on the left). The child also had a companion with them (hands visible on the right).

The robot can be seen signing in this video.


The practical outcome of this project was to prove that rapid prototyping can be used to bring high-end technology to those people who can benefit from it the most. Parents of the children felt it was extremely important that this type of research is being done.

The theoretical outcome of this project is the design framework for social robots, as well as the 5 design guidelines for the design of robots for children with autism.

The design framework which was developed in the thesis is available in full here.

The image below describes the basic idea of the framework: defining the problem space of the robot's design first, along with ethical considerations. This problem is then projected into the robot solution space, via design guidelines.

The design framework describes the process of designing social robots

The reasoning for the design guidelines for designing robots for autistic children are further explained in the thesis. They are based on research on autism, robots used with autistic children, as well as consultation with the speech therapist and psychologist. They are:


Behavior tree frameworking, creating a new design framework for social robots, rapid prototyping

Back to Portfolio