Archives

Modeling & Generating Routines

Leave a comment

Leveraging Human Routine Models to Detect and Generate Human Behaviors

An ability to detect behaviors that negatively impact people’s wellbeing and show people how they can correct those behaviors could enable technology that improves people’s lives. Existing supervised machine learning approaches to detect and generate such behaviors require lengthy and expensive data labeling by domain experts. In this work, we focus on the domain of routine behaviors, where we model routines as a series of frequent actions that people perform in specific situations. We present an approach that bypasses labeling each behavior instance that a person exhibits. Instead, we weakly label instances using people’s demonstrated routine. We classify and generate new instances based on the probability that they belong to the routine model. We illustrate our approach on an example system that helps drivers become aware of and understand their aggressive driving behaviors. Our work enables technology that can trigger interventions and help people reflect on their behaviors when those behaviors are likely to negatively impact them.

drivingsimulator_no_labelNikola Banovic, Anqi Wang, Yanfeng Jin, Christie Chang, Julian Ramos, Anind K. Dey, and Jennifer Mankoff. 2017. Leveraging Human Routine Models to Detect and Generate Human Behaviors. (To appear) In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (CHI ’17). ACM, New York, NY, USA.

3D Printing with Embedded Textiles

Leave a comment

screen-shot-2017-01-09-at-9-03-14-pm

Stretching the Bounds of 3D Printing with Embedded Textiles

Textiles are an old and well developed technology that have many desirable characteristics. They can be easily folded, twisted, deformed, or cut; some can be stretched; many are soft. Textiles can maintain their shape when placed under tension and can even be engineered with variable stretching ability.

When combined, textiles and 3D printing open up new opportunities for rapidly creating rigid objects with embedded flexibility as well as soft materials imbued with additional functionality. We introduce a suite of techniques for integrating the two and demonstrate how the malleability, stretchability and aesthetic qualities of textiles can enhance rigid printed objects, and how textiles can be augmented with functional properties enabled by 3D printing.

Click images below to see more detail:

Citation

Rivera, M.L., Moukperian, M., Ashbrook, D., Mankoff, J., Hudson, S.E. 2017. Stretching the Bounds of 3D Printing with Embedded Textiles. To appear in to the annual ACM conference on Human Factors in Computing Systems. CHI ‘17. [Paper]

Watch-ya-doin

Leave a comment
Watch-ya-doin is an innovative experienced based sampling framework for longitudinal data collection and analysis. Our system consists of a smartwatch and an android device working unobtrusively to track data. Our goal is to train on and recognize a specific activity over time. We use a simple wrist-worn accelerometer to predict eating behavior and other activities. These are inexpensive to deploy and easy to maintain, since battery life is a whole week using our application.
document3
     Our primary application area is AT abandonment. About 700,000 people in the United States have an upper limb amputation, and about 6.8 million face fine motor and/or arm dexterity limitations[1]. Assistive technology (AT), ranging from myo-electric prosthetics to passive prosthetics to a variety of orthotics can help in the rehabilitation and improve independence and ability to perform everyday tasks. Yet AT is not used to its full potential, with abandonment rates ranging from 23% to 90% for prosthetics users, and high abandonment of orthotics as well. Given the cost of these devices, this is an enormous waste of a significant financial investment in developing, fabricating, and providing the device, as well as potentially leading to frustration, insufficient rehabilitation, increased risk of limb-loss associated co-morbidities, and overall a reduced quality of life for the recipient.
       To address this, we need objective and accurate information about AT use. Current data is limited primarily to questionnaires, or skill testing during office visits. Apart from being limited by subjectivity and evaluator bias, survey tools are also not appropriate to estimate quality of use. A patient may – more or less accurately – report his or her AT use for a certain number of hours a day, but this does not indicate which tasks it was used for, which makes it difficult to evaluate how appropriate or helpful they were. In addition, neither reported use time nor skill testing can be sufficiently used to predict abandonment once AT is deployed.

      Our next steps include generalizing our approach to AT (such as upper limb prosthetics), and expanding it to include a wider variety of tracked activities. In addition, we will develop a longitudinal data set that includes examples of abandonment. This will allow the creation algorithms that can characterize the type and quality of use over the lifecycle of AT and predict abandonment.

[1] U.S. Census 2001

Printable Adaptations

Leave a comment

Reprise: A Design Tool for Specifying, Generating, and Customizing 3D Printable Adaptations on Everyday Objects

Reprise is a tool for creating custom adaptive 3D printable designs for making it easier to manipulate everything from tools to zipper pulls. Reprise’s library is based on a survey of about 3,000 assistive technology and life hacks drawn from textbooks on the topic as well as Thingiverse. Using Reprise, it is possible to specify a type of action (such as grasp or pull), indicate the direction of action on a 3D model of the object being adapted, parameterize the action in a simple GUI, specify an attachment method, and produce a 3D model that is ready to print.

Xiang ‘Anthony’ Chen, Jeeeun Kim, Jennifer Mankoff, Tovi Grossman, Stelian Coros, Scott Hudson (2016). Reprise: A Design Tool for Specifying, Generating, and Customizing 3D Printable Adaptations on Everyday Objects. Proceedings of the 29th Annual ACM Symposium on User Interface Software and Technology (UIST 2016) (pdf)

This slideshow requires JavaScript.

A Knitting Machine Compiler

2 Comments

 

A teddy bear wearing a knit hat, scarf (with pocket) and sweaterAlthough industrial knitting machines can automatically produce a wide range of garments, they are programmed through onerous means such as pixel level image manipulation. This limits the potential for automation of knitted object design, re-use of object components, and narrows the audience able to design for these machines. Our contribution is a visual design interface for specifying objects in terms of tubes and sheets and a compiler that can convert such an object into knittable machine instructions which handle knotty issues such as transfer planning (among needles) correctly. We demonstrate the range of objects our approach supports by example.

A Compiler for 3D Machine Knitting (SIGGRAPH 2016) Jim McCann, Lea Albaugh,
Vidya Narayanan, April Grow, Wojciech Matusik, Jennifer Mankoff, Jessica Hodgins

RapID — interactive RFID

Leave a comment

RapID – A framework for fabricating low-latency interactive objects with RFID tags

RFID tags can be used to add inexpensive, wireless, batteryless sensing to objects. However, quickly and accurately estimating the state of an RFID tag is difficult. In this work, we show how to achieve low-latency manipulation and movement sensing with off-the-shelf RFID tags and readers. Our approach couples a probabilistic filtering layer with a monte- carlo-sampling-based interaction layer, preserving uncertainty in tag reads until they can be resolved in the context of interactions. This allows designers’ code to reason about inputs at a high level. We demonstrate the effectiveness of our approach with a number of interactive objects, along with a library of components that can be combined to make new designs.

bestRapID: A Framework for Fabricating Low-Latency Interactive Objects with RFID Tags (CHI 2016, Page 5897) Andrew Spielberg, Alanson Sample, Scott E. Hudson, Jennifer Mankoff, James McCann

mixer-use quiz-use tic-tac-toe-play
pong-use pong-build tic-tac-toe-sketchup

Modeling Human Routines

Leave a comment

Modeling and Understanding Human Routine Behavior

Human routines are blueprints of behavior, which allow people to accomplish their purposeful repetitive tasks and activities. People express their routines through actions that they perform in the particular situations that triggered those actions. An ability to model routines and understand the situations in which they are likely to occur could allow technology to help people improve their bad habits, inexpert behavior, and other suboptimal routines. In this project we explore generalizable routine modeling approaches that encode patterns of routine behavior in ways that allow systems, such as smart agents, to classify, predict, and reason about human actions under the inherent uncertainty present in human behavior. Such technologies can have a positive effect on society by making people healthier, safer, and more efficient in their routine tasks.

Routines_Viz_Tool

Modeling and Understanding Human Routine Behavior
Nikola Banovic, Tofi Buzali, Fanny Chevalier, Jennifer Mankoff, and Anind K. Dey
In Proceedings of the 2016 ACM annual conference on Human Factors in Computing Systems(CHI ’16). ACM, New York, NY, USA.
Honorable Mention Award

Threadsteading

Leave a comment

 

In work done collaboratively with Disney Research Pittsburgh and led by Gillian Smith of Northeastern we explored a multi-player game that can be embedded into a quilting and/or embroidery machine interface. Gameplay is constrained by the fact that only a single thread of fabric can be drawn over time. Players compete to ‘scout’ over a map (a hex grid), where different hexes have different costs to explore.

Threadsteading map and custom control panel for quilting machine (light sunder fabric)

Threadsteading was accepted to Alt.CTRL.GDC.

G. Smith, A. Grow, C. Liu, L. Albaugh, J. Mankoff and J. McCann. Threadsteading: A single-line, two-player, territory-control game for quilting and embroidery machines. alt.ctrl.GDC 2016.

Dynamic question ordering

Leave a comment

In recent years, surveys have been shifting online, offering the possibility for adaptive questions, where later questions depend on responses to earlier questions. We present a general framework for dynamically ordering questions, based on previous responses, to engage respondents, improving survey completion and imputation of unknown items. Our work considers two scenarios for data collection from survey-takers. In the first, we want to maximize survey completion (and the quality of necessary imputations) and so we focus on ordering questions to engage the respondent and collect hopefully all the information we seek, or at least the information that most characterizes the respondent so imputed values will be accurate. In the second scenario, our goal is to give the respondent a personalized prediction, based on information they provide. Since it is possible to give a reasonable prediction with only a subset of questions, we are not concerned with motivating the user to answer all questions. Instead, we want to order questions so that the user provides information that most reduces the uncertainty of our prediction, while not being too burdensome to answer.

Publications
Kirstin Early, Stephen E. Fienberg, Jennifer Mankoff. (2016). Test time feature ordering with FOCUS: Interactive predictions with minimal user burden. In Proceedings of 2016 ACM Conference on Pervasive and Ubiquitous Computing. Honorable Mention: Top 5% of submissions. Talk slides.

3D printed attachments

Leave a comment

Encore: 3D printed attachments

What happens when you want to 3D print something that must interact with the real world? The Encore project makes it possible to 3D print objects that must attach to things in the real world. Encore provides an interface that, given an imported object and a chosen attachment method, visualizes metrics relating the goodness of the attachment. In addition, once an attachment type and location is chosen, Encore helps to produce the necessary support structure for attachment. Encore supports three main types of attachment: print-over, print-to-affix, and print-through.

Print-Over

Print-over attachments are printed directly on the existing object. This works well if the object is flat enough that the print head won’t encounter obstacles as it moves, and the object is made of a material that the printed material will easily adhere to. Encore helps by finding a rotation of the existing object that minimizes obstacles, and generating support material to hold the existing object in place.

Printing a magnet holder over a Teddy bear toy.
 
Left: printing an LED casing on a battery to make a simple torch; right: printing a handle to an espresso cup.
 

Print-to-Affix

An alternative that is useful when the existing object does not fit on the print bed is print-to-affix. In this approach, the attachment is designed to fit snugly against the existing object. It may be glued in place, or can include holes for a strap, such as a zip tie.

Left: printing a structure to make a glue gun stand; right: printing a reusable four-pack holder.
 

Print-Through

Finally, sometimes the attachment should be interlocked more loosely with the existing object. In this case, the process is to begin printing and stop the print partway through so that the existing object can be inserted. Encore can compute when this stopping point should be (and
whether it is possible)

A name tag printed through a pair of scissors
 
A bracelet printed through a charm
 

Encore the Design Tool

Encore is implemented in WebGL. It supports importation of an existing object, selection of an attachment, and then lets the user click to indicate where the attachment will go. Given this information, it uses geometric analysis to compute metrics for goodness of attachment, such as attachability and strength. Encore visualizes them using a heat map so that the user can adjust the attachment point.


Encore visualizes which parts of a wrench are more attachable when printing over a handle.

More Examples


Using print-to-affix to make a trophy from an egg holder
 

Using print-over to make a minion keychain
 

Using print-over to add a hanger to a screwdriver handle
 

Using print-through to make a key ring.
 

Using print-to-affix to make a battery case
 
Xiang ‘Anthony’ Chen, Stelian Coros, Jennifer Mankoff, Scott Hudson (2015). Encore: 3D Printed Augmentation of Everyday Objects with Printed-Over, Affixed and Interlocked Attachments. Proceedings of the 28th Annual ACM Symposium on User Interface Software and Technology (UIST 2015)