cs.ROOct 31, 2026

MESSPa: A Multi-modal Embodiment Stimulation Shoulder Pad for Preserving the Sense of Agency in a Detachable Robotic Arm

Authors: Vitvasin Vimolmongkolporn, Guillaume Gourmelen, Yukiko Iwasaki, Hiroyasu Iwata

Organizations: Department of Modern Mechanical Engineering, Waseda University, Tokyo, Japan

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Wearable robots such as supernumerary robotic limbs (SRLs) can augment human capability, but their effectiveness depends strongly on the operator's sense of embodiment, comprising the sense of agency (SoA) and the sense of ownership (SoO). When an SRL is detached from the body so that the operator can act in two places at once, both the sense of agency and task performance tend to degrade. This paper presents the Multi-modal Embodiment Stimulation Shoulder Pad (MESSPa), a wearable interface that combines one-to-one motion transfer with spatially distributed pneumatic (balloon) haptic feedback on the shoulder, with the goal of preserving the sense of agency across attached and detached states. We describe the system design, the control mapping, and a controlled experiment using a 2×22\times2 mixed design that crosses the control interface between groups (MESSPa, N=13N=13, vs. keyboard control, N=12N=12 from our previous study) and attachment (on-shoulder/detached) within participants, under single- and dual-task loads. Compared with keyboard control in the detached condition, MESSPa significantly reduced task completion time and significantly improved the sense of agency in the detached state, while narrowing the performance gap between the attached and detached conditions. The sense of ownership did not change significantly. These results indicate that combining motion transfer with positional haptic feedback is an effective way to recover agency when a robotic limb is operated remotely.

Figures & tables

Fig. 1: Illustration of MESSPa in use with the arm detached (experimental placement: Fig. 5 a). The HTC VIVE tracker on top of the shoulder pad maps the operator’s shoulder motion to the arm one-to-one, while four shoulder-mounted balloons deliver positional haptic feedback. The remote arm and the task screen are visible through the operator’s view.
Fig. 2: The MESSPa shoulder pad. Motion transfer is provided by an HTC VIVE tracker; haptic feedback is provided by four balloons (25 ×\times 25 mm each) arranged on the shoulder with 80 mm spacing, corresponding to the four quadrants of the arm’s workspace (Section III-C ).
Fig. 3: Pneumatic system. The compressor supply is regulated to 0.05 MPa by a proportional valve; four solenoid valves driven by an Arduino MEGA inflate the balloons according to the arm’s quadrant, as commanded by Unity.
Fig. 4: Balloon test bed used for force validation, consisting of a force gauge and a balloon locking mechanism that holds the balloon against the gauge while the input pressure is varied.
Fig. 5: (a) Experimental setup, showing the on-shoulder (attached) and detached conditions without MESSPa. (b) The Unity task screen, with the timer, the main laser-pointing target (green circle), and the sub-task colour indicator.
Sense of Agency (SoA)
A1It felt like I could control the robot arm as if it were my own arm
A2The movements of the robot arm were caused by my actions
A3I felt as if the movements of the robot arm were influencing my own movements
A4I felt as if the robot arm was moving by itself
Sense of Ownership (SoO)
O1I felt as if the robot arm was a part of my body
TABLE I: Sense of Embodiment Questionnaire (adapted from [ 27 ] ).
Fig. 6: Task completion time across the four wearing conditions, for the single and dual task, with the baseline from the previous experiment [ 15 ] . Brackets mark the three planned comparisons: ** p<0.01p<0.01 , * p<0.05p<0.05 (Bonferroni-adjusted), n.s. not significant.
Fig. 7: Error distance (mm) across the four wearing conditions, for the single and dual task. Brackets mark the three planned comparisons: ** p<0.01p<0.01 , * p<0.05p<0.05 (Bonferroni-adjusted), n.s. not significant.
Fig. 8: Sense of agency and sense of ownership in the detached condition, with and without MESSPa. Agency improves significantly with MESSPa ( p=0.017p=0.017 ); ownership is unchanged.
More in this topic

Robotics

Robotics studies the design and control of machines that perceive, move, and act in physical environments.

See all 768 papers

Explore similar work

CardsList

  1. Semantic Haptic Feedback Enhances Dexterous Robotic Teleoperation

    Aug 3, 2026Bingjian Huang, Sahar Aseeri, Jonas Schmidtler +9Robot TeleoperationHaptic Feedback

  2. Multimodal Embodied Agents: A Unified Capability-centric Perspective from Computer-Use to Robot-Use

    Oct 21, 2026Yanzhe Chen, Ziyi Yang, Jifeng Zhu +14Multimodal AgentsEmbodied AI

    Show Lab, National University of Singapore

  3. One Body, Two Minds: Variable Autonomy Approach for a Co-embodied Robotic Hand

    Jun 24, 2026Piotr Koczy, Yuchong Zhang, Danica Kragic +1RoboticsHuman-Robot Collaboration

    KTH Royal Institute of Technology, Stockholm, Sweden. · INCAR Robotics AB, Stockholm, Sweden.