|
|
© 1997 Copyright on this material is held by the authors.
Much ingenious work has been done in the area of tactile feedback, mostly in research labs and universities. For instance, the Z-Glove and Data-Glove [5] use gestures as input and tactile feedback as output.The PANTOGRAPH [3] provides the user with tactile and force feedback on a two dimensional plane. A mouse with tactile and force feedback [1], and joysticks using force feedback, ([2] and [4]) are examples of common input devices which have been enhanced to supply tactile feedback.
However, tactile feedback need not be the product of a complex and/or expensive research project. It can be easily implemented on commonly available platforms with equipment that is available at most electronics stores. In this paper we present a cheap, simple, versatile and effective tactile feedback device that can be built for less than $30 and requires no special software. It is usable on any computer with a parallel port, which covers virtually all Windows platforms.
Although it is a simple device, a relay can produce a variety of types of tactile feedback, ranging from single pulses to low frequency vibrations of a few hundred Hertz. There are other simple electrical devices, such as electric motors and piezo-electric buzzers, that can produce vibrations which we tried using in place of a relay. We found that a relay allows more precise control over the duration of tactile feedback than does an electric motor and is superior to a piezo-electric buzzer because it can vibrate at a variety of frequencies and can be used to generate short pulses. Further, as we were only interested in generating tactile feedback, we found it difficult to muffle the sound produced by the motor and piezo-electric buzzer when they were powered. By simply removing the protective casing of the relay and inserting cotton and masking tape over the contacts, we completely muffled any sound generated by the relay when it was powered. Note that any of these electronic devices may be directly substituted in the circuit for the relay: this makes experimentation with various kinds of tactile feedback, and sometimes aural feedback (as in the case of the piezo-electric buzzer), easy.
The relay must be placed so that a user can feel it vibrate. Since the protective plastic casing was removed from the relay, users could place their fingers directly on its contacts for greatest effect. An alternative is to attach the relay to a physical device that the user is grasping, such as a mouse, stylus or touchpad, and have the vibrations transmitted through the device. Because relays come in many shapes and sizes, they have great potential to be used in these ways.
The device has some drawbacks. After approximately 30 minutes of continuous usage, some users experience fatigue effects, including increased counting errors and slower counting speeds. As well, the "clicking" of the relay produces some unwanted sound feedback; thus, some care had to be taken to dampen the sound as we did not want sound feedback to be a confounding variable. This situation was easily rectified using the cotton and masking tape, but it should be noted that the additional auditory feedback may be advantageous for some, and perhaps most, applications.
We repeated our user test with sound feedback instead of tactile feedback, generating a "beep" rather than a pulse. A comparison between the two shows sound feedback is more intrusive than the tactile feedback, and users have to concentrate more on the feedback as a result. However, they are able to count beeps faster than pulses, which may be due to the characteristics of the aural vs. tactile senses, or to the method in which pulses are generated by our software. Each of these possibilities is currently being explored.
We believe there are many potential applications for this device. One, being investigated by a colleague, involves placing the relay under a touchpad. When the user puts pressure on the touchpad in order to perform the equivalent of a mouse click, the relay generates a pulse to provide tactile confirmation.
The ability of the relay to vibrate at various frequencies allows the implementation of tactile icons (tactile sensations that vary with the location of the tracker or state of the application). We could, for instance, cause the positioning device to vibrate when the user enters a mode, such as the object selection mode in a drawing program. Users would then be less likely to make the common mistake of inadvertently drawing lines because they think they are in object selection mode when they are in fact in line draw mode. Tactile icons can also be used instead of alert boxes (which use up screen real-estate and often must be explicitly cancelled) or sounds to inform the user of minor errors. Another application is in adapting visual interfaces for the blind. The relay can be attached to the mouse, stylus or touchpad being used for positioning and can emit a distinctive pulse whenever a blind user crosses over, for example, a menu boundary. This allows blind users to navigate through a pulldown menu by counting the number of pulses, which we have already proved feasible.
We have only begun to explore the types of tactile feedback possible with our device. We hope to improve the current mechanism by placing the relay inside the stylus itself, to test the effect of varying the frequency and amplitude of relay pulses, and to cheaply equip a mouse with tactile feedback by placing the relay inside the mousepad.
2. Balakrishnan, R., Ware, C., and Smith, T. Virtual Hand Tool with Force Feedback. Companion Proceedings of the CHI'94 Conference on Human Factors in Computing Systems, 83-84.
3. Ramstein, C. and Hayward, V. The PANTOGRAPH: A Large Workspace Haptic Device for Multi-Modal Human-Computer Interaction. Companion Proceedings of the CHI'94 Conference on Human Factors in Computing Systems, 57-58.
4. Rosenberg, L., and Brave, S. Using Force Feedback to Enhance Human Performance in Graphical User Interfaces. Companion Proceedings of the CHI'96 Conference on Human Factors in Computing Systems, 291-292.
5. Zimmerman, T. G., Lanier, J., Blanchard, C., Bryson, S. and Harvill, Y. A Hand Gesture Interface Device. Proceedings of ACM CHI + GI '87 Conference on Human Factors in Computing Systems and Graphics Interface, 189-192.
|
|