Robotic assisted design workflows: a study of key human factors influencing team fluency in human-robot collaborative design processes

Abstract

This paper presents the initial results and rationale towards the development of human-robot collaborative design workflows. An introduction to the basis of collaborative workflows and its impact for robots in architectural design is presented. Key elements, such as trust, reliance and robustness for the successful cooperation between humans and robots are identified and analysed. Human-robot collaboration is a multidimensional construct context dependent, this makes essential to understand how trust and team fluency develop when non-expert designers interact with industrial robots. A design process is then described based on sensor feedback, and phase-changing material formations that encourages human-robot collaboration during the genesis of the design. Two stages of development are presented. In stage one, an exploratory study was conducted to collect designers' opinions quantitatively and qualitatively. The results were analysed and led to the identification of the primary parameters that affect human-robot collaboration in the design process. In the second stage, machine learning is used to enhance the collaborative characteristics of the robotic partner in relation to the formation process of the material. The results reveal insights on human perceptions of robotic collaboration, and also explore neural-network-based feedback to enable expanded collaboration and communication between the robot and the designer.