HaptEx: Investigating Haptic Notification Channels for Exoskeletons Across Different Levels of Actuation

Vibrotactile Feedback & Skin StimulationForce Feedback & Pseudo-Haptic WeightHaptic WearablesPhysical Therapists & Rehabilitation SpecialistsAssistive Technology Specialists

Paper Title

HaptEx: Investigating Haptic Notification Channels for Exoskeletons Across Different Levels of Actuation

Publication Info

  • Topic area: Haptic feedback for human-exoskeleton interaction
  • Keywords: Haptic feedback, exoskeletons, proprioceptive notifications, thermal feedback, vibrotactile feedback, poking feedback, actuation levels, human-computer interaction, noticeability, urgency

Background and Problem

  • Problem / challenge: Existing studies on haptic feedback focus on low-movement contexts, leaving a gap in understanding how haptic notifications perform when the body is actively actuated by an exoskeleton. Sensory conflicts and masking effects may arise in such scenarios, affecting the reliability and noticeability of notifications.
  • Significance: Effective haptic notifications are critical for informing users of exoskeletons about system states, especially in safety-critical or high-stakes scenarios, without relying on visual or auditory channels.
  • Motivation and related work: Prior research has explored various haptic modalities for wearables but has largely neglected active contexts like those involving exoskeletons. This paper addresses the gap by systematically evaluating four haptic notification channels under different actuation levels.

Solution

  • Proposed approach: The study investigates the performance of four haptic notification channels (poking, proprioceptive, thermal, vibrotactile) integrated into a shoulder exoskeleton across three levels of actuation (none, moderate, strong).
  • Novelty:
    1. Systematic comparison of haptic notification channels under active exoskeleton actuation.
    2. Identification of the effects of actuation levels on noticeability, error rates, and response times.
    3. Insights into user preferences, urgency perception, and task interference for each channel.
    4. Design recommendations for selecting haptic channels based on urgency and actuation context.
  • Procedure and key techniques:
    • Participants performed a repetitive box-lifting task using a shoulder exoskeleton.
    • Notifications were delivered via four haptic modalities, and their noticeability, error rates, response times, and user experience were measured.
    • A within-subjects design was used, with counterbalanced conditions across 24 participants.
    • Quantitative data (e.g., response times, error rates) and qualitative feedback (e.g., interviews) were analyzed.

Results

  • Concrete findings:
    • Poking achieved the fastest response times (0.83–0.86 seconds) but had higher error rates compared to proprioceptive and thermal feedback.
    • Proprioceptive and thermal notifications had the lowest error rates (0–6%) and highest noticeability ratings.
    • Vibrotactile feedback had the highest error rates (47–71%) and was least noticeable, especially under actuation.
    • Actuation levels significantly affected error rates and noticeability but not response times.
  • Advantage over baselines:
    • Thermal feedback was most distinct and comfortable, avoiding sensory overlap with exoskeleton actuation.
    • Proprioceptive feedback conveyed the highest urgency but interfered most with task performance.
    • Poking was suitable for rapid alerts but suffered from occasional perceptual confusion.
  • Experiments / evaluation:
    • 24 participants (mean age: 26.3 years) completed 36 trials each, involving all combinations of four haptic modalities and three actuation levels.
    • Dependent variables included response times, error rates, and subjective ratings (e.g., noticeability, urgency, comfort).
  • Limitations and future work:
    • Limited to four haptic modalities and a single exoskeleton design.
    • Conducted in a controlled lab setting; real-world factors like ambient temperature and clothing were not tested.
    • Future work could explore additional modalities, richer notification encoding, and applicability to other exoskeleton designs and body locations.

Summary

This study systematically evaluates four haptic notification channels (poking, proprioceptive, thermal, vibrotactile) in a shoulder exoskeleton under varying actuation levels. Results show that thermal and proprioceptive notifications were most noticeable and accurate, with thermal feedback being the most comfortable and proprioceptive feedback conveying the highest urgency. Poking was fastest but prone to occasional errors, while vibrotactile feedback was the least effective under actuation. The findings highlight the importance of selecting haptic channels distinct from actuation-induced sensory interference. Recommendations include using thermal and vibrotactile feedback for non-urgent notifications and poking or proprioceptive feedback for urgent alerts. These insights provide a foundation for designing effective human-exoskeleton interactions.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/222808/2026

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3772318.3791179
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2026
emoji_events
Award
No award tagged
group
Authors
4 authors
sell
Subtopics
Vibrotactile Feedback & Skin Stimulation, Force Feedback & Pseudo-Haptic Weight, Haptic Wearables
work
Professions
Physical Therapists & Rehabilitation Specialists, Assistive Technology Specialists
article
Content Status
Full text indexed
hub
Related Papers
4 related papers