Making learning easy for children with autism using robots

The triadic interaction between educator, learner and robot in this pilot study showed promise

Updated - July 09, 2022 10:42 pm IST

Published - July 09, 2022 07:10 pm IST

The toy robot, Cozmo, would help to assist learning in children.

The toy robot, Cozmo, would help to assist learning in children. | Photo Credit: Special Arrangement

Teaching children with Autism Spectrum Disorder poses many challenges for special educators working in this area. One of these challenges is that the interventions needed are highly specific to each child. In a class, therefore, when handling more than one child, the teacher may benefit from support offered by robotic assistants — the commercially available robots. There is a need to research this aspect and provide cost-effective solutions.

This is the area of work in a pilot study carried out in the lab of Chiranjib Bhattacharya of Indian Institute of Science (IISc), Bengaluru, with the support of Jayashree Ramesh of the Academy for Severe Handicaps and Autism (ASHA), Bengaluru. Dr. Bhattacharya and his team studied the use of toy robots — small in size and not necessarily humanoid — in assisting learning for children with autism spectrum disorder. Their work and results have been published in a paper in 2021 IEEE International Conference on Robotics and Automation.

AI and learners

With the aim of understanding the role of Artificial Intelligence (AI) and toys in the learning process of children with neurodevelopmental challenges, the team approached ASHA, for a collaboration. “The founder, Ms. Jayashree Ramesh, was very supportive. She helped us identify which children to work with, the challenges to target and what their education is like,” says Nabanita Paul, a PhD student with Chiranjib Bhattacharya and one of the first authors of the paper. 

One part of the work was to study the effectiveness of interventions using a toy robot, Cozmo; the other part was to try and use drones to help children with motivational issues — for instance, motivating them to exercise if they were not inclined per se to do so (for instance, with hypotonia).

They started the experiments with Cozmo in 2019. In 2020, due to the lockdown, they had to try online experiments. They started with seven children to be taught, and at the end of the experiments, they had to manage with four children. 

The goals were threefold — what it takes to understand the interaction between the special educator, child and robot; focus on helping not only the child but also the special educator In reducing their workload; designing the interactions and evaluation metrics.

The interventions were designed to be in keeping with the Individualised Education Plan (IEP) that was in place for each child,  comprising not only formal education goals but also special needs such as social communication or motor skills. As an example of an IEP, there could be a simple goal, Ms. Paul explains, “If you want to make a drawing and you need a pencil, you need to ask, ‘please give me a pencil’. This could be a challenge for these children.” This kind of functional communication can include asking for things, asking for help, asking to use the washroom, etc.  

Lesson plan

So the lessons were designed with this kind of outcome in mind. One lesson was “Talk to Me” where the child and the robot would talk to each other and practice self-introduction questions. “For example, Cozmo would ask the child, ‘what is your name?’ If the child did not respond, the special educator would prompt the child, or the robot would also prompt the child, saying, ‘I didn’t hear you’,” says Ms. Paul. “If the child answered partially or incorrectly, the robot would make a little jig or show excitement. The special educator would prompt the child,” she adds. 

To estimate their success, the researchers would measure, for example, the number of prompts given by the special educator in the presence and absence of the robot Cozmo. 

Since this was a pilot study, they had just five participants. “Some of the children showed great interest in the robot sessions, they even started small spontaneous conversations with me — a total stranger,” says Ms. Paul.

Award and citation

The work has already seen the limelight, and Ms. Paul has won the Murthy Govindaraju “Women in Computer Science” endowment award 2022, which comprises a cash prize of ₹1,00,000 and a citation to the awardee. The award has been instituted by Madhukar C. Govindaraju, alumnus of IISc, and his wife, Chaya Murthy Govindaraju, to promote excellence of women in computer science and to advance the computer science area among young professionals, especially women students.

0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in


Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.