This brilliant work is described in the paper “LineFORM: Actuated Curve Interfaces for Display, Interaction, and Constraint “
Partly inspired by biomimetic designs for Snake-like robots, this work focuses on interaction with the “snake”, and use as a computer interface.
The device is a chain of motors, each capable of folding in 1D (i.e., as a hinge). The chain also has sensors to detect the position of the joints, to enable direct manipulation of the shape.
Nakagaki and colleagues show that such chains can represent not only lines, but also surfaces and 3D volumes, and also can execute movements in 2D and 3D.
These strings can be used for several kinds of display, including Data Representations, Iconic Forms, User Interface Elements (e.g., buttons), Ergonomic and Aesthetic Form, Tactile and Haptic Display. See the video for illustrations of these!
The chain is also a tactile input, capable of sensing touch, push, pull, pinch, etc., and of being stretched, tied in knots, and so on.
The LineFORM is also, “wearable”, and can function to “constrain” motion, as a body-worn prosthetic, or as a tool.
Very cool, and you should read the paper.
As in all great work, this project makes me think. (That is my highest compliment.)
How it moves. As the authors note, the quick snap-into-shape motion can be surprising and possibly alarming for a person, especially when the LineFORM is fairly large and close (relative to the human). I would guess that both the speed and the spikiness are key to the human reaction to the motion. In addition, the apparent autonomy of the device may cause alarm. How can the LineFORM signal its intentions? E.g., that it is about to snap into shape, or that it is about to do something on its own.
Indeed, this may even be a safety factor, if the human needs to release the LineFORM to avoid injury or interference with proper function.
I note that the current device is sometimes linked to a computer screen, which provides opportunities to signal the user. But for many scenarios, the computer will be interacting only via the LineFORM, so there must be a way for it to safely and gently tell the human to expect motion. Perhaps we need a design vocabulary of ‘snakey machine gestures’.
(Thinking about this, I certainly would love to see user errors or ambiguous inputs signaled by the LineFORM rearing up and forming a “?” at you! :-))
The authors are focuses on the direct manipulation interface, which is cool. They note that it is difficult to create some forms, such as wire meshes. This is both a limitation of the specific ‘string’,, and a limitation of input methods. Just as it is difficult to hand draw a wire mesh, it is difficult to ‘hand bend’ one.
So—there could be a way to load subroutines into the LineFORM. This could be used to save and share hand made shapes, but also to algorithmically generate shapes that can be ‘down loaded’ into the LineFORM. Sort of like loading fonts or game prefabs, only 3D spatial configurations.
Finally, this demo made me thing about using this concept at different scales.
Scaling up, imagine a LineFORM at the size of furniture. This could be a really, really, reconfigurable sofa/chair/bed/etc. It would also be capable of actively reconfiguring, upon command, or in response to context, or even as a safety or comfort feature. (I guess we should consider the possibilities for erotic play, while we’re at it.)
Clearly, at this scale, we need to be careful about how the device moves, and about the ‘haptics’ of the surfaces and the resistance and push of the actuators.
Scaling down, I’m thinking about wearables. Think something like active fringes on a jacket. They can act as augmentation, giving an extra hand or temporary pocket to hod things. They can deliver messages or signal events tactilely. They can also act independently, expressing emotional or social meaning.
Imagine that me and my partner both have jackets with ‘smart fringes’ that dance together when we are near! In fact, our jackets could embrace, and touch and exchange private messages.
Cool stuff! I need one!
- Nakagaki, Ken, Sean Follmer, and Hiroshi Ishii, LineFORM: Actuated Curve Interfaces for Display, Interaction, and Constraint, in Proceedings of the 28th Annual ACM Symposium on User Interface Software & Technology. 2015, ACM: Daegu, Kyungpook, Republic of Korea. p. 333-339.