Improved VT Feedback Mappings [87/131]

Improved VT Feedback Mappings [87/131]
- Introduction [12]
  - Situated at the overlap of Neuroscience (what can be perceived) and Engineering (what can be rendered).
    - + Intersection with Domain
    - (Research groups very limited)
    - here Physiotherapy/motor learning
  - and Study at overlap of what can be sensed/displayed and physically reached (task)
  - Introduce VT Feedback [5]
    - shown to be effective in
      - sense of texture perception
      - supporting motor learning
      - virtual reality immersion
        
        multimodal sensory integration
      - providing localized (timing) cues
  - Identify Gap [2]
    - While there has been progress towards wearable VT FB (Wall/Sienko) and mass-market tactor tech (Lofelt/FB), the specific application/resarch domain remains a chicken/egg situation
    - Low low-level 'jamming', as shown for spatiotemporal patterns (VibeViz), would seem logical for the feedback case
    - Progression
      - einfache signalgebung / symbolisch
        
        tacton
        
        nct
        
        transparent
      - interactiv/feedback / subsymbolisch
        
        Posture
        
        Balance
        
        belt
        
        Guidance
  - Outline Approach [3]
  - Chapter Overview [2]
- Background [26]
  - Biofeedback and Augmented Feedback [2]
    - Definitions
    - Modalities and Applications
    - Case Study Example we consider: VT Balance Belts
      - established research + rehabilitation tool
      - obvious location for balance and orientation feedback
      - for in-home rehabilitation
        
        stroke survivors
        
        elderly
        
        with corona, telemedecine rises
      - slow amelioration of VT haptics
        
        single actuator displays
        
        vibration motors ftw
  - Sense of Touch [5]
    - Physiological Basis
    - Comparison with Auditory Perception
    - Perceptual Dimensions
    - Tactile Illusions
    - often part of multimodal perception
    - may be uniqely suited in situations where eyes and ears are not available.
      - elderly and impaired
      - just even listening to music/talking
      - other dualtasking
  - Vibrotactile Actuators [4]
    - Evolution and Overview
    - Eccentric rotating mass
    - Voice coils
    - Tactor Drivers
    - Other Vibrotactile Actuator Types
  - Wearables [4]
  - Sensing Postural Parameters [3]
    - Optical Motion Capture
    - MARG (Magnetic, Angular Rate and Gravity) Sensors
    - Pressure Based Sensing
    - Other Sensors
  - Embedded Computing [3]
    - Microcontroller Architectures
    - Prototyping Systems
    - Communication
  - Overview of VTW Applications [5]
    - Notifications
    - Tactons
    - Medical
    - Entertainment
    - Control
      - Virtual switches
    - Design tools
    - Gaming
- Analysis: State of the Art []
  - Continuous Vibrotactile Feedback on Posture
  - Review Topics
    - "Vibrotactile wearable Feedback"
      - since 2022
        
        lofelts showing up in the wild
        
        really horrible bike guidance prototype
        
        some good review papers
  - Shortcomings of the literature [2]
    - More refined stimuli not systematically investigated
    - Complex Stimuli in Perception Research Decades ago
    - ERM vs VC <-> control mechanism
    - Chicken & Egg Situation with Toolkits
    - Latency not investigated
    - Existing research on complex tactile stimuli can be grouped into two main lines of study:
      Continous and in isolation for building a model of the pacinian channel, and as tactons, short structured tactile notifications.
  - What applications do we want to support? [2]
- Approach []
  - Goal
    - Align perceptual ability with with current common tactor's rendering capability
    - Goal is to provide groundwork for improved systems, basic science rather than applied, going into the perceptual real
    - Contribution should be a solid basis, demonstrating science craftmanship and laying foundation for future work
  - Challenge
    - so many variables in mapping
    - nervous system is adaptive, shadowing bad interface design and the need for bio-inspired interface design
    - interesting research is proposed in design-centered conferences such as CHI; but rarely ties back to neuro-science side of it
    - neuroscience-side is decades old sometimes (early 2000s)
  - Questions
  - Approach: Improving on the SoA by combining
- Modular VT Toolkit [10]
  - Goal
    - feedback systems and general purpose wearable building blocks
    - Situate yourself for qualitative Understanding of the Problem
      - enable in-situ studies
    - 80/20 solutions
      - can we do w/o stationary sensors?
      - need to understand physics of interaction
        
        distance sensor for shoe worked great
      - bno055 just good enough, was thousands of dollars of heavy hw 15 years ago
    - Open & Modular
      - rapidly changing options
        
        constant change/evolution
        
        tech just gets better and better
        
        enabling applications that were just not possible before
      - less is more / modular is key
        
        the best part is no part
      - learnability + knowledge transfer
      - accessible and compatible
    - haptics as first class citizen
      - more actuators
        
        shoe
        
        belt
        
        nct
      - better actuators
        
        low-frequency voice coil actuators
        
        now available
        
        before: 250 Hz domination :(
        
        comfort
        
        expressivity
- VT Belt
  - Apparatus
    - Belt HW Evolution [5]
      - mkI [2]
      - mkII [3]
      - mk III [3]
  - Baseline Questions (technical limits)
    - This part is well underway
    - Quantification of tactor response limits?
    - base-line comparison
      - freehanging 100 gram load.
      - baseline: Static
      - with erm first (also for latency)
    - Pulsing
      - Frequency response
      - Shortest pulse length
      - Highest pulse frequency
      - "Step Response"
    - ERM: Open Loop vs Closed
      - ~\cite{Rotella2012}?
    - LoFelt
      - Stimulus Parameters
        
        Base Freq (68 Hz?)
        
        Pulse Freq (Range?)
        
        Pulse Shape (Square, Sine, Pulse)
        
        Pulse Width (for Square)
    - System Latency
  - Novel Questions (perceptual limits)
    - This part is where I am stuck
    - Are ERMs, with suitable control, capable of coming close to the more expensive voice coil tactors, or does the Lofelt's higher temporal resolution enable measureable performance improvements?
      - Is higher temporal resolution of the broadband actuators reflected in decreased error?
      - comparison erm-lofelt with amplitude control
    - Is Pulsing beneficial?
      - maybe more physically plausible response than static with varying intensity
      - tactile sense bw-limited -> rumbling rather than static
      - needs to be low freq to not be audible
      - NIW needed hf auditory part to be convincing
      - on the timescale of tactile sensation and physical movement
    - Which posture-vibration mappings are most effective?
      - Maybe we need to start with something like multidimensional scaling to determine proper stimuli parameter scales?
      - Activation strength vs. pulse vs combined?
      - What makes a stimulus more comfortable, which mappings are more and less fatiguing, and other subjective / qualitative aspects communicated by users
      - Improved task performance: time to target / trajectory length / error / attempts / trajectory around target
      - Deadzone?
    - Mapping Aspects
      - Types of Error
        
        targets/min
        
        Time-to-Target
        
        Stabilogram: Fit ellipse of sway trajectory around targets
        
        Angular Error
      - Qualitative:
        
        stimuli comfort
        
        Rating (Intuitiveness of conditions)
      - Feedback
        
        Feedback proportional to distance to target (== error)
        
        deadband around target - only deadzone showed short-term aftereffects \cite{Ballardini2020}
        
        Repulsive or attractive feedback? Will try both in pilot; users seem to not care (~\cite{Rotella2012}) or prefer repulsive (Sienko), but attractive seems to elicit a faster movement to target.
        
        tried but discarded:
        
        saltation for z-axis/yaw rotation possible? good? or roughness chase?~\cite{Bensmaia2005}
    - How does latency affect perfomance?
- Study 1 [8]
  - General Structure
    - Explain, Consent (10 min)
    - 6 Blocks à 3 Minutes
      - Resting position
        
        Zero IMU, Height
      - n targets in sequence
      - Resting Position
      - Repeat
      - Break, switch belt, recalibrate (2min)
    - Debrief
      - Questionnaire
    - Sanitize
- Study 2 [8]
- Discussion [5]
- Conclusion [2]
- Appendix
  - A: Auditory Feedback for Slacklining [2]
  - B: Vibrotactile Shoes [15]
    - Background: Plantar Vibrotactile Systems
      - Tactile Perception on the Feet
      - Single-Actuator / Stationary Systems
      - Multi-Actuator Shoes and Insoles
      - Spatiotemporal Patterns: Tactons
      - Gait Cueing / Parkinson’s
      - Environmental Synthesis
    - Goals and Requirements
      - Personas
      - Intended Use Cases
      - Requirements
    - General Design
    - Validation of Placement: Prototype 1, Tethered
      - Vibrotactile Actuator Choice
      - Actuator Placement
      - Prototype Haptic Sandal
      - Tacton Discrimination: Sitting vs. Standing
    - Discussion
    - Improved Coupling and Integration for Locomotion: Prototype
    - Goals and Requirements
    - Improved Actuator Integration
    - Untethered System Architecture
    - Discussion
  - C: Vibrotactile Shoe Evaluation [20]
    - Tacton Discrimination: Standing vs. Walking
      - Introduction
      - Tactons
      - Study
      - Results
      - Discussion
      - Conclusion and Outlook
    - Effects of different feedback modalities on radial menu selection
      - Introduction
      - Related Work
      - Methodology
      - Experiment Design
      - Results and Discussion
      - Conclusion and Outlook
      - Discussion and Outlook
      - Stochastic Resonance and Sub-Threshold Balance Aids
      - Energy Harvesting
      - Conclusion
      - Acknowledgments
      - Publications and Disclosures
  - D: Further Applications of the Modular Toolkit
    - Brain-Computer Interfaces for Minimally Responsive Patients [5]
      - Requirements / Goals
        
        "Situated Observations" (Qualitative)
        
        bci communicator
        
        motor training
        
        modular prototyping can help in exploring the parameter space
      - System Architecture
        
        Gameboi developed according to needs of clinical experimenters
      - Sensors and Tactors on Hands: Vibrotactile Glove
    - Neumann-Cosel System [2]