This study addresses the communication challenges faced by nonspeaking autistic individuals by developing an Augmented Reality (AR)-based system that reduces reliance on human assistants. Using Behavioural Cloning, the system learns from real-life interactions to personalise the placement of a virtual letterboard based on each user’s motor patterns. The approach improves accuracy over non-ML methods, offering a promising step toward more autonomous and private communication for nonspeakers.
This study addresses the communication challenges faced by nonspeaking autistic individuals by developing an Augmented Reality (AR)-based system that reduces reliance on human assistants. Using Behavioural Cloning, the system learns from real-life interactions to personalise the placement of a virtual letterboard based on each user’s motor patterns. The approach improves accuracy over non-ML methods, offering a promising step toward more autonomous and private communication for nonspeakers.
