ProjecTA: A Semi-Humanoid Robotic Teaching Assistant with In-Situ Projection for Guided Tours

Paper Title

ProjecTA: A Semi-Humanoid Robotic Teaching Assistant with In-Situ Projection for Guided Tours

Publication Info

• Topic area: Human-robot interaction in educational and nomadic learning environments.
• Keywords: Robotic teaching assistant, in-situ projection, nomadic learning, cognitive load, multimodal coordination, gesture-based interaction, makerspace, guided tours, user engagement, human-computer interaction.

Background and Problem

• Problem / challenge: Current robotic teaching assistants (TAs) rely on body-mounted screens, which can distract learners from physical objects, increase cognitive load, and hinder shared access to information in nomadic learning contexts. There is limited empirical research on integrating in-situ projection with robotic TAs to address these issues.
• Significance: Reducing cognitive load and improving multimodal coordination in learning environments can enhance learner experiences and outcomes, particularly in spatially distributed, object-centered settings like makerspaces.
• Motivation and related work: Prior work has demonstrated the benefits of in-situ projection and gesture-based interaction in fixed-location tasks but has not explored their integration with mobile robotic TAs in nomadic learning. This study builds on these findings to address gaps in empirical comparisons and design implications for such systems.

Solution

• Proposed approach: ProjecTA, a semi-humanoid robotic TA equipped with a head-mounted projector, gesture-capable arms, and a mobile chassis, designed to deliver guided tours in makerspaces by coordinating in-situ projection, gestures, and speech.
• Novelty:
  1. Development of ProjecTA, a robotic TA integrating in-situ projection with gestures and speech for nomadic learning.
  2. Empirical comparison of ProjecTA with a screen-based counterpart, revealing significant reductions in extraneous cognitive load and enhanced user experiences.
  3. Derivation of design implications for robotic TAs leveraging in-situ projection in physical learning environments.
• Procedure and key techniques:
  • ProjecTA uses a presentation choreography workflow to align speech, visuals, and gestures on a unified timeline.
  • The system employs a head-mounted projector for near-object overlays, gesture modules for deictic, iconic, and metaphoric gestures, and a mobile chassis for navigation.
  • A mixed-methods study compared ProjecTA with a screen-based baseline in a university makerspace, involving 24 participants across two guided tour rounds.

Results

• Concrete findings:
  • ProjecTA significantly reduced extraneous cognitive load (Mdiff = −2.15, p < 0.001, Cohen’s d = −1.086).
  • Higher perceived usability (Mdiff = 0.60, p < 0.01, d = 0.692) and multimodal complementarity (Z = −2.958, p < 0.01, r = 0.604).
  • No significant differences in intrinsic load, quiz performance, aesthetic appeal, or focused attention.
• Advantage over baselines:
  • ProjecTA outperformed the screen-based counterpart in reducing attention switching, improving shared access to information, and enhancing the coordination of visuals, gestures, and speech.
• Experiments / evaluation:
  • Within-subject design with 24 participants in a university makerspace.
  • Measures included cognitive load (CLS), user engagement (UES-SF), quiz performance, and semi-structured interviews.
  • Counterbalanced tours covered six pieces of equipment, with randomized modality order.
• Limitations and future work:
  • Lack of mixed-initiative interaction; future work could enable Q&A and dialogue.
  • Occasional image jitter and projection distortions; future iterations could improve hardware stability.
  • Limited generalizability to other contexts; future studies could explore diverse learning environments and larger participant cohorts.

Summary

ProjecTA introduces a novel robotic teaching assistant that integrates in-situ projection with gestures and speech to support nomadic learning in physical environments. Empirical results demonstrate its ability to significantly reduce extraneous cognitive load and enhance user experiences compared to a screen-based counterpart. By reducing attention switching and improving multimodal coordination, ProjecTA offers a promising approach for guided tours in makerspaces and similar settings. Design implications and future directions highlight opportunities for refining projection placement, multimodal orchestration, and extending applications to other collaborative learning scenarios.