Towards Human-Chatbot Interaction: A Virtual Assistant for the Ramp-up Process

Abstract

In this paper, a grounding framework is proposed that combines unsupervised and supervised grounding by extending an unsupervised grounding model with a mechanism to learn from explicit human teaching. To investigate whether explicit teaching improves the sample efficiency of the original model, both models are evaluated through an interaction experiment between a human tutor and a robot in which synonymous shape, color, and action words are grounded through geometric object characteristics, color histograms, and kinematic joint features. The results show that explicit teaching improves the sample efficiency of the unsupervised baseline model.

Hey Max, can you help me? An Intuitive Virtual Assistant for Industrial Robots

“…In the area of teaching and learning, Prgorskly et al [22] proposed a virtual learning assistant as an assessment tool to support online learners' self-regulation in online learning, while Preece et al [23] investigated the effectiveness of a controlled natural language-based conversational interface for human-machine interaction regarding locally observed collective knowledge. In the manufacturing domain, Barbosa et al [24] created a VA, which supported the training process of the technical personnel in industrial plant operators, Casillo et al [25] built a chatbot for the training of employees in an Industry 4.0 scenario and Zimmer et al [26] developed a VA to assist the ramp-up process of an assembly system. In our previous work, we have presented a VA named Bot-X to assist the workers in handling a variety of complex manufacturing tasks, e.g., order processing, and production execution [27].…”

Hey Max, can you help me? An Intuitive Virtual Assistant for Industrial Robots

“…In the area of teaching and learning, Prgorskly et al [22] proposed a virtual learning assistant as an assessment tool to support online learners' self-regulation in online learning, while Preece et al [23] investigated the effectiveness of a controlled natural language-based conversational interface for human-machine interaction regarding locally observed collective knowledge. In the manufacturing domain, Barbosa et al [24] created a VA, which supported the training process of the technical personnel in industrial plant operators, Casillo et al [25] built a chatbot for the training of employees in an Industry 4.0 scenario and Zimmer et al [26] developed a VA to assist the ramp-up process of an assembly system. In our previous work, we have presented a VA named Bot-X to assist the workers in handling a variety of complex manufacturing tasks, e.g., order processing, and production execution [27].…”

Hey Max, can you help me? An Intuitive Virtual Assistant for Industrial Robots

“…In the manufacturing domain, Casillo et al [40] proposed a chatbot for the training of employees in an Industry 4.0 scenario. Barbosa et al [41] created a VA, which can support the training process of the technical personnel in industrial plant operators for real-time operations, while Zimmer et al [42] developed a VA to assist the ramp-up process of an assembly system. In our previous work, we have presented a VA named Bot-X to assist the workers in handling a variety of complex manufacturing tasks, e.g., order processing, and production execution [43].…”