Chinese

2019,  1 (2):   163 - 175   Published Date：2019-4-20

DOI: 10.3724/SP.J.2096-5796.2019.0011

Abstract

Background
Our previous studies have shown that electrical stimulation from the skin surface to the tendon region (Tendon Electrical Stimulation: TES) can elicit a force sensation, and adjusting the current parameters can control the amount of the sensation. TES is thought to present a proprioceptive force sensation by stimulating receptors or sensory nerves responsible for recognizing the magnitude of the muscle contraction existing inside the tendon, so it can be a proprioceptive module of a small-size, low-cost force feedback device. But there is also suspect that TES presents only strong, noisy skin sensation. From previous study, it was found that TES has some limitation on varying sensations.
Methods
In this study, in addition to characterizing the proprioceptive sensation induced by TES, we constructed a multimodal presentation system reproducing a situation in which force is applied to the hand was offered, so as to investigate whether TES contributed to the reproduction of haptics cooperating with other modalities, rather than disturbing them. Specifically, we used vibration to present a cutaneous sensation and a visual head mounted display (HMD) system to present simultaneous images. Using this system, we also evaluated the efficacy of TES itself and that of the multimodal system involving TES.
Results
We found that TES, along with visual and vibration stimulation, contributed to the perception of a certain force.
Conclusions
Thus, TES appears to be an effective component of multimodal force sense presentation systems.

Content

1 Introduction
In many Virtual Reality (VR) systems, haptic feedback can improve the sense of immersion or operability. For instance, in medical trainings in VR, force feedback enables the user to feel softness of organs so as to avoid hurting them. On the other hand, most haptic devices to date have the common problem of requiring a large hardware setup to mechanically generate actual force.
Our previous literature demonstrated that a force sensation could be generated by electrical stimulation of the tendon via the skin surface (referred to here as tendon electrical stimulation or TES), which can be presented with a small setup. We also showed that magnitude of the sensation could be controlled by adjusting the current parameters[1] (Figure 1). Unlike the presentation of a force sensation via electrical muscle stimulation (EMS), which is used frequently in computer human interactions [2,3,4], TES stimulates sensory nerves, not motor nerves. Thus, TES can be used to present a force sensation without generating unnecessary muscle contractions.
However, the sensation produced by TES has several limitations. First, while TES can be used to present a constant force, it cannot present a momentary sensation, such as that associated with a collision. Second, interpretation of the sensation varies among users. In a previous experiment (described in section 3), when the dorsal part of the wrist was stimulated, most participants reported that the direction of the elicited force sensation was from the back of the hand to the palm (flexion direction). However, when participants were asked to freely comment on the sensation, some revealed that they had felt as if their hand was bent at the wrist, while others felt that their forearm was bent at the elbow, or that their entire arm was pushed.
In this paper, we attempted to address these issues by giving multiple cues to the user. Specifically, we sought to integrate multiple sensation presentation modalities related to force sensation.
Force sensation is often integrated and perceived through multiple modalities. Cutaneous sensation and proprioception are especially important[5,6,7,8], and the role of visual sensation is also considered to be large[9].
In this study, we constructed a multimodal presentation system that simultaneously performs cutaneous stimulation and visual stimulation in addition to TES (Figure 2). We investigated how the force sensation differed between multimodal stimulation and unimodal stimulation (i.e., TES only).
2 Related work
Common haptic devices that can present both proprioce-ptive and cutaneous sensations are roughly classified into body-grounded (exoskeleton or wearable type) and environment-grounded devices. While these devices generally require a large number of motors for multi-degrees of freedom actuation, a number of recent methods have been proposed to induce force sensation at a relatively low cost.
In this section, we describe systems that can elicit a sense of force with a relatively small mechanical setup.
2.1 Pseudo-haptics with visual presentation
Pseudo-haptic presentation is a convenient way to present a force sensation without a motor[9]. For instance, a haptic sensation can be induced by changing the speed of a cursor on a screen that is providing visual feedback regarding the movement of a mouse operated by a user[10]. Pusch et al. induced pseudo haptics by exposing the hands of users to a visual force field[11].
2.2 Force sensation presentation involving cutaneous sensation
Cutaneous sensations include various types such as pressure and vibration. The cutaneous sensation associated with pressure and shearing force plays a major role in force sensation, and various methods to induce this sensation producing skin deformation, have been proposed[12,13,14,15].
Vibration sensation can also be used for force feedback. Amemiya et al.[6] found that asymmetric vibration could induce a force-like sensation, and this phenomenon is now receiving much attention[16,17,18]. Vibration is an important cue in the sensation of collision, and damped oscillations have been proposed to model and reproduce this sensation[19].
2.3 Force sensation presentation involving proprioception
Proprioception includes sensations processed by receptors, such as muscle spindle and the Golgi tendon organ (GTO) in skeletal muscles or vestibular organs. One method for eliciting proprioception that has a relatively simple mechanical setup is Electrical Muscle Stimulation (EMS), in which the muscle acts as an actuator and produces an actual force[2,3,4,20,21]. EMS-induced force is realized by causing the contraction of an antagonist muscle located on the opposite side of a skeletal muscle that is the target of the force sensation. One drawback of EMS-based methods is the requirement of activity in the antagonist muscle, which would not actually occur in a real situation. Users feel an external force, but they also feel their muscle activity.
In TES, instead of muscle contractions, the sensory nerves in or around the tendons connected to the target skeletal muscles are stimulated to present a force sensation. However, as mentioned in the introduction, the interpretation of TES varies among users, which necessitates its use in combination with other modalities.
3 Multimodal haptic presentation system adopting TES
3.1 Methods for presenting a force sensation via TES and the direction of the sensation
In our previous TES literature[1], electrical stimulation was applied via electrodes placed mainly on the surface of the skin directly above the tendon at the back of the wrist. The electrical current was controlled while applying electrical stimulation. The current was a rectangular wave with a frequency of 80Hz, a pulse width of $200\mathrm{\mu }\mathrm{s}$ , and a pulse height of about 10mA on average, with a maximum value of 25mA. The stimulus was intended to produce a sensation of force travelling in the direction opposite the side to which the electrode was affixed.
The mechanism of generation of the force sensation is thought to be derived from the response of deep receptors to electrical stimulation of the tendons. We conducted preliminary experiments in which we varied the electrode intervals. Specifically, we compared trials with narrow-interval electrodes (which were expected to stimulate only a shallow portion, or skin) to those with wide-interval electrodes (which were expected to stimulate both shallow and deep part). We observed that the force sensation was more clearly generated in the latter case, implying that deep receptors were involved in the sensation. Since tendons, and not muscles, were mainly present at the stimulation site, the GTO was strongly suspected to be the stimulated receptor involved in the elicited sensation. Since GTO’s activity means the force magnitude of the muscle contraction, if the GTO is the responsible receptor, TES modulates magnitude of the force sensation. Also, since TES does not ignite motor nerves, the posture of the wrist does not change. Therefore, the users are expected to perceive an illusory external force travelling opposite the direction of the perceived muscle contraction.
3.2 Multimodal Force Sensation Presentation System
The system, which consists of an electrical stimulation device[22], a vibrator (Acouve Lab., Vibro transducer Vp 2 series Vp 210), and a head mounted display (Oculus, DK 2), presents sensations in three modalities: visual, vibratory, and force (Figure 2, Figure 3). These modules are integrated by a PC. The electrical stimulation device and transducer are controlled by microcontrollers (NXP, mbed LPC 1768), which synchronize information with the PC via serial communication. We used two gel electrodes (1.9mm × 3.5mm, Vitrode F-150S, Nihon Kohden Corporation) for TES.
3.2.1 Visual stimulation via a head mounted display
We constructed a Virtual Reality (VR) space on the PC, such that the Head Mounted Display (HMD) basically always displays a model right hand with the thumb facing upward and the hand open (Figure 4). A virtual object moves and collides with the backside of the hand. The speed of the object when it contacts the hand is 0.8m/s. The object keeps pushing the hand for 3s after contact. During that period, the right hand slightly moves in the direction of the palm (0.2m) as if pushed by the object, and then stays in place.
3.2.2 Mechanical stimulation by vibrator
It is difficult to present the sensation of an impact using TES. In our TES model, a cutaneous sensation is presented via the vibrator. The input to the vibrator is controlled by amplifying a sinusoidal wave generated by a microcontroller. The microcontroller is communicating with the microcontroller that controls TES, which is sensitive to timing within the μs range. This allows the delay of the vibration to be within a few μs of the electrical stimulation. We used Vp2 (Acouve Lab., Vibro transducer Vp2 series Vp210) as a vibrator in the experiment. The diameter of the vibrator was 43mm. In the Experiment, it was put on the back of the hand so that the direction of the vibration was perpendicular to the surface of the back of the hand. The input voltage signal to the vibrator was decaying sinusoidal wave that is often used for the expression of collision[19,23], defined as following: $V\left(t\right)=A{e}^{-Bt}\mathrm{s}\mathrm{i}\mathrm{n}\mathrm{}\left(2\mathrm{\pi }ft\right)$ [V]; where $A$ was the amplitude, the only controllable parameter by an amplifier’s volume in the experiment, $B\approx 10.0$ , and $f=100\left[\mathrm{H}\mathrm{z}\right]$ , which were defined in a preliminary test by the authors (Figure 5).
3.2.3 　 Electrical stimulation over a tendon
TES expresses a static continuous force, performed continuously while the object is in contact with and pushing the right hand. The electrodes are positioned on the back of the wrist, as in the previous experiment[1], and two electrodes are placed so as to straddle the wrist joint (Figure 6). For the electrical stimulation, we used a biphasic rectangular pulse wave with a frequency of 100Hz, which is decided as a parameter that produce clear sensation in preliminary trials referring to the previous literature, and a pulse width of 200μs, which is definitely same to the parameter of the literature. The two electrodes alternately acted as the stimulating electrodes. Moreover, it is possible to adjust the height of the pulse to the extent that the participants can sufficiently perceive the force sensation up to 25mA.
4 Experiment
4.1 Experiment preparation
Figure 2 shows an overview of the experiment. The participants wore the HMD, the electrodes on the back of the wrist of the right hand, and the vibrator on the back of the right hand, which we selected as a natural position considering the TES-generated force sensation. The pulse height of the electrical stimulation and the amplitude of the vibration were adjusted in advance by the participants. The electrical stimulation was adjusted so that the participant could clearly feel the force sensation by the TES only, and the vibration was adjusted so that they sufficiently feel the impact sensation.
4.2 Experimental conditions
The experimental conditions are as shown in Table 1. We used six conditions in total, excluding two conditions "only vibration stimulus" and "no stimulus" from the eight combinations of TES, vibration, and visual presentation. Among them, the “only TES” condition was used as the reference stimulus and the other five conditions were presented once randomly as a comparison stimulus. That is, each participant performed five trials. The reference stimulus was presented immediately prior to each comparative stimulus.
Experimental Condition. E: TES. M: Mechanical stimulation by vibration. V: Visual stimulation by HMD
4.3 Procedure
The participants were instructed to match the posture of their right hand as best as they could to the right-hand model in the VR space. In the VR space, the object appeared at the right side of the screen at a random time, and made a uniformly accelerated motion to the left. When the object collided with the hand model, mechanical vibratory stimulation and electrical stimulation were presented. In trials where no visual stimulus was presented, the object and the right-hand model disappeared, and the collision and subsequent force sensation involved only the mechanical vibration stimulation and electrical stimulation.
In experiment, the participants were asked to respond to questions about the following four items:
(1) Magnitude of the force sensation
(2) Naturalness of the force sensation
(3) Direction and rotational axis of the force sensation
(4) Point of application of the force sensation
Item (1) and (2) were asked to evaluate comparison conditions compared with the reference condition, while item (3) and (4) were to characterize all the six conditions. Therefore, prior to all trials, the participants were given a reference stimulus and asked about the latter two items ((3) and (4)) regarding the reference condition; after that, in each trial, the participants were given a reference stimulus followed by a comparison stimulus, then were asked to respond to questions about all the four items regarding the comparison condition.
Item (1) and (2) were scored on a 7-level Likert scale, where 4 corresponded to the reference stimulus. If the induced sensation was not perceived as a force sensation, we asked participants to rate the value as 1.
Item (3) was answered by selecting from seven options: the hand bent inward (flexion) or outward (extension) at the wrist, the forearm bent inward or outward at the elbow, the entire arm pushed inward or outward, or none of the above (Figure 7).
Item (4) was answered by selecting from seven options: on the palm, on the back of the hand, on the palm side of the wrist, on the back of the wrist, on the palm side of the arm, on the back of the arm, or none of the above.
The naturalness was intended to be answered as a cue showing how well the stimulations go along with each other in direction or position of force sensation, or how better it was felt as a unified sensation.
4.4 Participants
Eleven participants 22 to 24 years of age (10 males, all right-handed) were recruited in the laboratory. They know the notion of haptic feedback, but they did not have experience of electrical stimulation-elicited force sensation. One of them was excluded from data analysis because he declared that he had made a mistake in his responses.
5 Result
5.1 Magnitude of the force sensation
As shown in Figure 8a, conditions without TES always produced a lower score compared with the condition with only TES (reference condition), whereas conditions with TES always produced the same or higher scores compared with the condition with only TES. The results for each stimulus were compared to those for the reference stimulus using Wilcoxon's sign test. We found significant differences in the magnitude of force sensation under the condition with only the visual stimulus, the condition with the visual + mechanical stimuli, and the condition with all the three modalities (significance was denoted by p≤0.05). That is, under all conditions in which TES was not presented, the magnitude of the force sensation was significantly smaller than that of the reference stimulus, and under the condition in which all of the modalities were presented in combination, the force sensation was significantly larger compared with that of the reference stimulus.
5.2 Naturalness of the force sensation
The results regarding the naturalness of the force sensation are shown in Figure 8b. Under all of the conditions other than the condition in which the visual stimulus was presented alone, the force sensation was generally perceived as equally as or more natural as/than that in the reference stimulus. The results for each stimulus were compared to those for the reference stimulus using Wilcoxon's sign test. We found a significant difference in the naturalness of the force sensation under the condition with the visual stimulus alone and that with all three modalities (significance was denoted by p≤0.05). That is, the naturalness in the condition in which only the visual stimulus was presented was significantly reduced with respect to the reference stimulus, and the naturalness under the condition in which the stimulus was presented through all the modalities was significantly higher than that in the reference stimulus.
5.3 Direction and rotational axis of force sensation
The results regarding the direction and rotational axis of the force sensation are shown in Figure 9a. All of the participants answered that at least the force sensation was felt to occur in an inward direction, that is, the direction that the palm was facing, regardless of the axis. "None of the above" was chosen mainly under conditions in which TES was not presented. Under conditions in which TES was presented, there was a tendency for participants to answer that the force sensation with which the forearm moved inward had the elbow as the origin of movement. When stimulating with three modalities, there were an almost equal number of responses for each of the following options: around the wrist, around the elbow, and perpendicular to the surface of the forearm and upperarm.
5.4 Position of the point of application of the force sensation
The results regarding the position of the point of application are shown in Figure 9b. In nearly every trial, the participants answered that the application points were on the back of any part of the arm. There were some comments that the point of action was inside the arm, or that the sensation was like being pulled rather than being pressed. When the visual stimulus or/and the mechanical stimulus was/were given, there was a tendency for participants to answer that the point of application was on the back of the hand. Especially under the condition with both the visual and mechanical stimuli, the position of the point of application was mostly reported to be perceived on the back of the hand. However, under the condition with presentation of all three modalities, this rate decreased.
6 Discussion
Our experimental results confirmed that the magnitude and naturalness of a force sensation was significantly increased by the simultaneous presentation of information via three modalities compared with only electrical stimulation to the tendon (reference stimulus). However, detailed control of the direction and axis of the force sensation and the position of the applied force will require further characterization. The findings obtained from the different trial types are listed below.
6.1 Magnitude of the force sensation
The experimental results showed that the combination of all three stimuli significantly increased the magnitude of the perceived force compared with the condition in which only the TES was presented. In contrast, the magnitude of the force sensation was significantly smaller in trials in which only the visual stimulus was presented and trials just with the visual and mechanical vibratory stimuli.
The latter result suggests that TES greatly contributed to the generation of the force sensation, since all conditions without TES had a smaller magnitude of force sensation compared with the reference. The former result can be explained by the increase in realism due to the simultaneous presentation of multi-modal sensations; however, we cannot exclude the possibility that participants judged the amount of force symbolically (i.e. they might have interpreted more types of sensation as a stronger sensation). We will discuss this in the following section.
6.2 Naturalness of the force sensation
Compared with the reference condition, naturalness significantly decreased under the vision only condition, and significantly increased under the tri-modal condition. Therefore, the three modalities did not interfere with one another, and rather, cooperated in the presentation of information of the same situation. This supports the idea, mentioned in the previous section, that the amount of perceived force increased due to heightened realism.
The results regarding magnitude and naturalness indicate that TES is effective as a method of presenting force sensations, and that it can present more natural and convincing force sensations in combination with other modalities.
6.3 Direction and axis of the force sensation
Most participants reported that the direction of the force sensation was at least from the back of the hand to the palm (when around the wrist, this equals flexion, not extension). This reconfirms that TES is distinct from EMS. If motor nerves were stimulated and the force sensation was generated by muscle contraction, the force should have been felt in the extension direction, not flexion, since the electrodes were placed on the backside of the wrist.
Our interpretation of the direction of the force is as follows. The TES stimulated the tendon of the muscles, which elicited the sensation that the extensor was contracting. Since the wrist was not actually extended, users interpreted that some external force was working to hinder the extension of the wrist. In other words, they felt that they were competing against the force that was trying to refract the wrist (Figure 10). Based on this interpretation, the reports could have agreed even on the point that the direction was around the wrist.
In the case of the TES only trials, however, the majority of participants reported that they felt the direction of the force to be inward (from the backside to the palm side) around the elbow, not the wrist. This might be because the position of force application was ambiguous, and turning the elbow might have been a more natural behavior for users than turning the wrist.
If so, application of a vibration to the back of the hand should reduce this ambiguity, and might make participants believe that the point of application is on the back of the hand such that the force refracts their wrist. In our trials, when a vibration was presented without TES, this tendency was observed, but when TES was given with a vibration, many participants reported that the rotational force was around the elbow. This might be because the TES lasted for about 3s, and therefore dominated the vibration, which lasted only 0.5s. Moreover, the TES also elicits a cutaneous sensation around the wrist that might have affected the interpretation.
6.4 Point of application of the force sensation
Under the reference condition, or the TES-only condition, many participants reported that the point of application was at the wrist. This might be due to the associated cutaneous sensation caused by electrical stimulation. Under the bi-modal condition in which vision and vibration were both presented, participants predominantly responded obediently that the force was applied at the back of the hand. When combining the vibratory stimulation and the electrical stimulation were combined, a nearly equal number of participants reported that the force was applied at the back of the wrist and the hand, although as mentioned, we expected that it was possible to convince users that the point of application is at the back of the hand by presenting a vibration. Thus, the cutaneous sensation caused by the electrical stimulation may have dominated the experience for some people.
6.5 Other findings
During the experiments and demonstrations held on some conferences, we found that although the presentation of mechanical vibration and electrical stimulation were synchronous, about 10% of the participants commented that they started feeling the electrical stimulation after the end of the vibration. When they took off the vibrator and experienced the stimulation again, they commented that the collision in the visual scene and the start of electrical stimulation appeared to be synchronous certainly. This means that as the start of the electrical stimulation was not physically delayed, the perception of electrical stimulation might have been masked by mechanical vibration.
We also observed that a few participants moved their hands at the time of stimulus presentation as if actual object had pushed their hand. Several of them said that they thought this was a kind of reflex. This was seen even when no mechanical vibration was given but TES was conducted. This reconfirms our assumption that the TES stimulated sensory nerves, not motor nerves; because if the muscle was stimulated by electrical stimulation, the movement should have been in the opposite direction to the observed movement.
6.6 Future work
Our final goal is to reproduce a complex haptic sensation by combining the presentation of various textures and proprioceptive force sensations. Towards this, we hope to refine the TES method to induce proprioception without presenting a cutaneous sensation. This will clarify the roles of cutaneous and proprioceptive presentation.
TES is characterized by the presentation of a force without controlling the posture of the body. However, in the case of a virtual reality scenario where users are allowed to engage in arbitrary motion, the users’ hand may pass through a virtual object after they make intentional contact. To address this issue, we plan to apply tendon vibration stimulation, which has been found to successfully elicit a sense of position that is different from the actual position[24], or partially use EMS.
Whether the generation of force sensation via TES is actually due to the contribution of proprioceptive organs, particularly Golgi tendon organs, is unclear, although a preliminary experiment suggested that nerves located deeper than the skin surface may contribute to the effect. In future work, we will attempt to elucidate the mechanisms underlying the force sensation induced by TES using neurophysiological methods.
As we have mentioned in the introduction, our aim is to use TES as an alternate for force-feedback device, which generally require space and high cost. Although we showed TES around wrist part, in principle TES can induce force sensation around any joint where a tendon isn’t overlapped by muscles: for instance, elbows, and ankles. We consider it is particularly suitable for situations in which users’ motion space is limited, such as in a cockpit, since TES does not induce physical movement contrary to the other force-feedback devices.
7 Conclusion
We examined two issues in tendon electrical stimulation. One is that the sensation elicited by TES cannot express momentary situations like a collision, and the other is that the interpretation of the sensation varies among users. To tackle these issues, we constructed a multimodal system that combines vision, cutaneous sensation, and proprioception. Using this system, we examined how the force sensation changed with respect to the magnitude of the force sensation, the enhancement of naturalness, and the control of the direction and the position of force application. As a result, we confirmed that the magnitude and naturalness of the force sensation significantly increased under the tri-modal condition compared with the TES-only condition. Therefore, TES appears to be a sufficiently effective method for presenting a force sensation. Future work will address issues such as the fine control of the direction of the force sensation and the position of the point of application.

Reference

1.

Takahashi A, Tanabe K, Kajimoto H. Relationship Between Force Sensation and Stimulation Parameters in Tendon Electrical Stimulation. In. Singapore: Springer Singapore, 2018, 233–238 DOI:10.1007/978-981-10-4157-0_40

2.

Tamaki E, Miyaki T, Rekimoto J. PossessedHand: techniques for controlling human hands using electrical muscles stimuli. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. Vancouver, BC, Canada, ACM, 2011, 543–552 DOI:10.1145/1978942.1979018

3.

Lopes P, Ion A, Baudisch P. Impacto: simulating physical impact by combining tactile stimulation with electrical muscle stimulation. In: Proceedings of the 28th Annual ACM Symposium on User Interface Software & Technology. Charlotte, NC, USA, ACM, 2015, 11–19 DOI:10.1145/2807442.2807443

4.

Tamaki E, Chan T, Iwasaki K. UnlimitedHand: input and output hand gestures with less calibration time. In: Proceedings of the 29th Annual Symposium on User Interface Software and Technology. Tokyo, Japan, ACM, 2016, 163–165 DOI:10.1145/2984751.2985743

5.

Minamizawa K, Prattichizzo D, Tachi S. Simplified design of haptic display by extending one-point kinesthetic feedback to multipoint tactile feedback. In: 2010 IEEE Haptics Symposium, 2010, 257–260 DOI:10.1109/HAPTIC.2010.5444646

6.

Amemiya T, Maeda T. Asymmetric oscillation distorts the perceived heaviness of handheld objects. IEEE Transactions on Haptics, 2008, 1(1): 9–18 DOI:10.1109/toh.2008.5

7.

Ebied A M, Kemp G J, Frostick S P. The role of cutaneous sensation in the motor function of the hand. Journal of Orthopaedic Research, 2004, 22(4): 862–866 DOI:10.1016/j.orthres.2003.12.005

8.

Prattichizzo D, Pacchierotti C, Rosati G. Cutaneous force feedback as a sensory subtraction technique in haptics. IEEE Transactions on Haptics, 2012, 5(4): 289–300 DOI:10.1109/toh.2012.15

9.

Pusch A, Lécuyer A. Pseudo-haptics: from the theoretical foundations to practical system design guidelines. In: Proceedings of the 13th international conference on multimodal interfaces. Alicante, Spain, ACM, 2011, 57C64 DOI:10.1145/2070481.2070494

10.

Lécuyer A, BurkhardtJ-M, Etienne L. Feeling bumps and holes without a haptic interface: the perception of pseudo-haptic textures. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. Vienna, Austria, ACM, 2004, 239–246 DOI:10.1145/985692.985723

11.

Pusch A, Martin O, Coquillart S. HEMP-hand-displacement-based pseudo-haptics: a study of a force field application. In: 2008 IEEE Symposium on 3D User Interfaces. 2008, 59–66 DOI:10.1109/3DUI.2008.4476593

12.

Minamizawa K, Kajimoto H, Kawakami N, Tachi S. A wearable haptic display to present the gravity sensation-preliminary observations and device design. In: Second Joint EuroHaptics Conference and Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems. 2007, 133–138 DOI:10.1109/WHC.2007.15

13.

Matsue R, Sato M, Hashimoto Y, Kajimoto H. “Hanger reflex”: A reflex motion of a head by temporal pressure for wearable interface. In: 2008 SICE Annual Conference, 2008, 1463–1467 DOI:10.1109/SICE.2008.4654889

14.

Kuniyasu Y, Sato M, Fukushima S, Kajimoto H. Transmission of forearm motion by tangential deformation of the skin. In: Proceedings of the 3rd Augmented Human International Conference. Megève, France, ACM, 2012, 1–4 DOI:10.1145/2160125.2160141

15.

Yem V, Vu K, Kon Y, Kajimoto H. Effect of electrical stimulation haptic feedback on perceptions of softness-hardness and stickiness while touching a virtual object. In: 2018 IEEE Conference on Virtual Reality and 3D User Interfaces (VR). 2018, 89–96 DOI:10.1109/VR.2018.8446403

16.

Amemiya T, Gomi H. Active manual movement improves directional perception of illusory force. IEEE Transactions on Haptics, 2016, 9(4): 465–473 DOI:10.1109/toh.2016.2587624

17.

Rekimoto J. Traxion: a tactile interaction device with virtual force sensation. In: Proceedings of the 26th annual ACM symposium on User interface software and technology. St. Andrews, Scotland, United Kingdom, ACM, 2013, 427–432 DOI:10.1145/2501988.2502044

18.

Culbertson H, Walker J M, Raitor M, Okamura A M. A wearable asymmetric vibration excitation system for presenting three-dimensional translation and rotation cues. In: Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems. Denver, Colorado, USA, ACM, 2017, 4972–4982 DOI:10.1145/3025453.3025741

19.

Okamura A M, Cutkosky M R, Dennerlein J T. Reality-based models for vibration feedback in virtual environments. ASME Transactions on Mechatronics, 2001, 6(3): 245–252 DOI:10.1109/3516.951362

20.

Lopes P, Baudisch P. Muscle-propelled force feedback: bringing force feedback to mobile devices. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. Paris, France, ACM, 2013, 2577–2580 DOI:10.1145/2459236.2459276

21.

Pfeiffer M, Duente T, Rohs M. Let your body move: a prototyping toolkit for wearable force feedback with electrical muscle stimulation. In: Proceedings of the 18th International Conference on Human-Computer Interaction with Mobile Devices and Services. Florence, Italy, ACM, 2016, 418–427 DOI:10.1145/2935334.2935348

22.

Kajimoto H. Electrotactile display with real-time impedance feedback using pulse width modulation. IEEE Transactions on Haptics, 2012, 5(2): 184–188 DOI:10.1109/toh.2011.39

23.

Wellman P, Howe R D. Towards realistic vibrotactile display in virtual environments. In: Proceedings of the ASME Dynamic Systems and Control Division. 1995, 57(2): 713–718

24.

Goodwin G M, McCloskey D I, Matthews P B C. The contribution of muscle afferents to keslesthesia shown by vibration induced illusionsof movement and by the effects of paralysing joint afferents. Brain, 1972, 95(4): 705–748 DOI:10.1093/brain/95.4.705