Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 1. Complex activities such as playing a musical instrument present a great challenge to upper limb amputees. Tasks such as these require accurate slip control of multiple fingertips simultaneously across different surfaces. In this paper, we explore the potential for three amputees and nine non- amputees to simultaneously control the state of sliding contact at two fingertips simultaneously by integrating two channels of variable frequency vibrotactile haptic feedback into their motor control strategies. The rendering of the man in this image was licensed and modified for public display.

Article Snippet: The Slow and Fast vibrotactile activation signals for this test were created in Simulink using the ROS toolbox.

Techniques: Control, Modification

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 4. Characterization of the two vibration modes of the vibrotactile stimulators for haptic feedback to the human subjects from the index (I) and little (L) fingers. (A) The BioTac SP on the I finger of the Shadow Hand was used in this experiment to measure the Slow and Fast vibration modes of the vibrotactile stimulators that was conveyed to the human subjects. (B) Steady–state pressure (PDC) and (C) spectrogram of the steady-state pressure (PDC) measured by the BioTac corresponding to the Slow vibration mode. (D) The dynamic pressure signal (PAC) and the (E) corresponding spectrogram from the Slow vibration mode. (F) Steady–state pressure (PDC) and (G) spectrogram corresponding to the Fast vibration mode. (H) The dynamic pressure signal (PAC) and (I) the corresponding spectrogram from the Fast vibration mode.

Article Snippet: The Slow and Fast vibrotactile activation signals for this test were created in Simulink using the ROS toolbox.

Techniques:

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 5. Training subjects for efferent control. (A) The amplified, filtered, and rectified EMG signals from six electrodes were normalized. (B) ANN classifier outputs for the Sim, I, L, and NM classes corresponding to the six EMG signals. (C) Subject S3 performing the EMG classifier training. The red and blue circles indicate where the vibrotactile stimulators for the index (I) and little (L) fingers were respectively placed for subject S3. The subject (S3) gave permission for the use of his image.

Article Snippet: The Slow and Fast vibrotactile activation signals for this test were created in Simulink using the ROS toolbox.

Techniques: Control, Amplification

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 6. Control system for multichannel sensorimotor integration. Six EMG signals were classified by subject-specific ANNs to specify which finger(s) the subjects wanted to control. The EMG signals were also used to specify the desired forces for the index (I) and little (L) fingers that were realized by hybrid force–velocity controllers. ANNs were used with the BioTac SPs on the I and L fingers to classify the 24 taxels at each fingertip into sensations of sliding contact, either up or down. These sensations of touch from each fingertip were encoded via the frequency of vibration and fed back to the subjects with the haptic armband. The red and blue circles indicate where the vibrotactile stimulators for the index (I) and little (L) fingers were respectively placed for subject S1. EMG electrodes were located under the black armband. The subject (S1) gave permission for the use of his image.

Article Snippet: The Slow and Fast vibrotactile activation signals for this test were created in Simulink using the ROS toolbox.

Techniques: Control

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 7. Sample data illustrating robotic system operation for the four cases of simultaneous slip at the index (I) and little (L) fingers. Simultaneous haptic feedback had two different vibration frequencies depending upon the direction of sliding contact. Upward slip was encoded with Slow vibration, while the downward slip produced Fast vibration. Subjects were trained to permit upward slip but prevent downward slip using their six EMG signals to produce four different classes (Sim, I, L, NM) with the efferent ANN. (A) Simultaneous downward slip at each fingertip was created by (B) both stepper motors driving downward slip. (C) This caused both vibrotactile actuators to be activated in the Fast mode. (D) The subject perceived the simultaneously activated channels of haptic feedback and increased his EMG signals (E) to produce the Sim class with the efferent ANN. (F,G) Slip down at the I finger with slip up at the L finger caused the (H) vibrotactile stimulators to be actuated with the Fast and Slow modes, respectively. (I) The subject responded to this multichannel

Article Snippet: The Slow and Fast vibrotactile activation signals for this test were created in Simulink using the ROS toolbox.

Techniques: Produced

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 1. Complex activities such as playing a musical instrument present a great challenge to upper limb amputees. Tasks such as these require accurate slip control of multiple fingertips simultaneously across different surfaces. In this paper, we explore the potential for three amputees and nine non- amputees to simultaneously control the state of sliding contact at two fingertips simultaneously by integrating two channels of variable frequency vibrotactile haptic feedback into their motor control strategies. The rendering of the man in this image was licensed and modified for public display.

Article Snippet: The haptic feedback training process started by demonstrating the Slow and Fast vibrotactile stimulator activation modes through Simulink in real time.

Techniques: Control, Modification

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 4. Characterization of the two vibration modes of the vibrotactile stimulators for haptic feedback to the human subjects from the index (I) and little (L) fingers. (A) The BioTac SP on the I finger of the Shadow Hand was used in this experiment to measure the Slow and Fast vibration modes of the vibrotactile stimulators that was conveyed to the human subjects. (B) Steady–state pressure (PDC) and (C) spectrogram of the steady-state pressure (PDC) measured by the BioTac corresponding to the Slow vibration mode. (D) The dynamic pressure signal (PAC) and the (E) corresponding spectrogram from the Slow vibration mode. (F) Steady–state pressure (PDC) and (G) spectrogram corresponding to the Fast vibration mode. (H) The dynamic pressure signal (PAC) and (I) the corresponding spectrogram from the Fast vibration mode.

Article Snippet: The haptic feedback training process started by demonstrating the Slow and Fast vibrotactile stimulator activation modes through Simulink in real time.

Techniques:

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 5. Training subjects for efferent control. (A) The amplified, filtered, and rectified EMG signals from six electrodes were normalized. (B) ANN classifier outputs for the Sim, I, L, and NM classes corresponding to the six EMG signals. (C) Subject S3 performing the EMG classifier training. The red and blue circles indicate where the vibrotactile stimulators for the index (I) and little (L) fingers were respectively placed for subject S3. The subject (S3) gave permission for the use of his image.

Article Snippet: The haptic feedback training process started by demonstrating the Slow and Fast vibrotactile stimulator activation modes through Simulink in real time.

Techniques: Control, Amplification

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 6. Control system for multichannel sensorimotor integration. Six EMG signals were classified by subject-specific ANNs to specify which finger(s) the subjects wanted to control. The EMG signals were also used to specify the desired forces for the index (I) and little (L) fingers that were realized by hybrid force–velocity controllers. ANNs were used with the BioTac SPs on the I and L fingers to classify the 24 taxels at each fingertip into sensations of sliding contact, either up or down. These sensations of touch from each fingertip were encoded via the frequency of vibration and fed back to the subjects with the haptic armband. The red and blue circles indicate where the vibrotactile stimulators for the index (I) and little (L) fingers were respectively placed for subject S1. EMG electrodes were located under the black armband. The subject (S1) gave permission for the use of his image.

Article Snippet: The haptic feedback training process started by demonstrating the Slow and Fast vibrotactile stimulator activation modes through Simulink in real time.

Techniques: Control

Journal: Robotics

Article Title: Multichannel Sensorimotor Integration with a Dexterous Artificial Hand

doi: 10.3390/robotics13070097

Figure Lengend Snippet: Figure 7. Sample data illustrating robotic system operation for the four cases of simultaneous slip at the index (I) and little (L) fingers. Simultaneous haptic feedback had two different vibration frequencies depending upon the direction of sliding contact. Upward slip was encoded with Slow vibration, while the downward slip produced Fast vibration. Subjects were trained to permit upward slip but prevent downward slip using their six EMG signals to produce four different classes (Sim, I, L, NM) with the efferent ANN. (A) Simultaneous downward slip at each fingertip was created by (B) both stepper motors driving downward slip. (C) This caused both vibrotactile actuators to be activated in the Fast mode. (D) The subject perceived the simultaneously activated channels of haptic feedback and increased his EMG signals (E) to produce the Sim class with the efferent ANN. (F,G) Slip down at the I finger with slip up at the L finger caused the (H) vibrotactile stimulators to be actuated with the Fast and Slow modes, respectively. (I) The subject responded to this multichannel

Article Snippet: The haptic feedback training process started by demonstrating the Slow and Fast vibrotactile stimulator activation modes through Simulink in real time.

Techniques: Produced

Journal: Science robotics

Article Title: Controlling sensation intensity for electrotactile stimulation in human-machine interfaces

doi: 10.1126/scirobotics.aap9770

Figure Lengend Snippet: For every condition, subjects were asked to match a reference sensation intensity at three differing values of Rp. (A) Peak pulse energy (Ep) and (B) phase charge (Q) for a subject across 2 magnitudes of sensation, 3 stimulation locations, and 2 electrode sizes using electroconductive gel to change Rp and an electrotactile reference. (C) Ep and (D) Q for a subject with a below-elbow amputation (Subject TR1) at a weak magnitude of sensation over the right biceps using a vibrotactile reference and either electroconductive gel or exercise to change Rp. In (A)-(D), initial pre-gel/exercise values of Ep and Q vs. Rp (blue) are used to compute lines of constant sensation intensity (dashed lines). When Rp changes, the controller computes new stimulation parameters to stay on the lines of constant sensation intensity (red). At the controller-computed pulse duration, subjects adjusted the current amplitude to match a constant reference sensation intensity, and we derive Ep and Q (green). (E) R2 regression statistics from fitting the controller-computed lines of constant sensation intensity to the subject-derived values of Ep and Q across 10 subjects without arm impairment in Exp. 2A (10 subjects × 5 conditions, n = 50) as well as 9 subjects without arm impairment and Subject TR1 in Exps. 2B-2C (10 subjects × 1 condition, n = 10). The R2 distributions are shown in (F) for Ep vs. Rp and (G) for Q vs. Rp. A sign-rank test indicated that the R2 values are statistically significantly greater than 0.7 (p < 0.05).

Article Snippet: Instead, a vibrotactile motor (310–103, Precision Microdrives, London, UK) was used to provide a reference sensation intensity over the left biceps, under the assumption that changes in skin impedance have little to no effect on vibrotactile stimulation intensity.

Techniques: Derivative Assay

Journal: Scientific Reports

Article Title: Biased visuospatial perception in complex regional pain syndrome

doi: 10.1038/s41598-017-10077-8

Figure Lengend Snippet: Illustration of the experimental set-up for the tactile TOJ task. Tactile stimuli are generated by two vibrotactile transducers held between the thumb and the index finger of each hand. The hands are either held in an uncrossed posture ( a ) or crossed over the sagittal body midline ( b ). Participants are blindfolded and presented with white noise. The figure depicts a participant affected by left-sided CRPS.

Article Snippet: Vibrotactile stimuli were generated by two vibrotactile transducers driven by standard audio amplifiers (TL-002–14R Haptuator Redesign, Tactile Labs, Inc., Montreal, Canada).

Techniques: Generated