A nonlinear contextually aware prompting system (N-CAPS) to assist workers with intellectual and developmental disabilities to perform factory assembly tasks: system overview and pilot testing

Abstract

Purpose: This paper presents a new cognitive assistive technology, nonlinear contextually aware prompting system (N-CAPS) that uses advanced sensing and artificial intelligence to monitor and provide assistance to workers with cognitive disabilities during a factory assembly task. Methods: The N-CAPS system was designed through the application of various computer vision and artificial intelligence algorithms that allows the system to track a user during a specific assembly task, and then provide verbal and visual prompts to the worker as needed. A pilot study was completed with the N-CAPS solution in order to investigate whether it was an appropriate intervention. Four participants completed the required assembly task five different times, using the N-CAPS system. Results: The participants completed all of the trials that they attempted with 85.7% of the steps completed without assistance from the job coach. Of the 85.7% of steps completed independently, 32.5% of these were completed in response to prompts given by N-CAPS. Overall system accuracy was 83.3%, the overall sensitivity was 86.2% and the overall specificity was 82.4%. Conclusions: The results from the study were positive in that they showed that this type of technology does have merit with this population.

Implications for Rehabilitation

• It provides a concise summary of the importance of work in the lives of people with intellectual disabilities and how technology can support this life goal.

• It describes the first artificially intelligent system designed to support workers with intellectually disabilities.

• It provides evidence that individuals with intellectual disabilities can perform a work task in response to technology.

Keywords:

• Artificial intelligence
• intellectual disabilities
• job coaching
• pilot study
• prompting
• technology
• workplace

Acknowledgements

The authors gratefully acknowledge all study participants and coordinators who helped in the preparation and completion of the described pilot study.

Declaration of interest

None of the authors have any competing interests to declare. This project was funded by the RERC for the Advancement of Cognitive Technologies through a research grant from the National Institutes of Disability and Rehabilitation Research (NIDRR) under the US Department of Education Grant #H133E090003.