living with ubiquitous sensing and dynamic responsive media
TRANSCRIPT
Prof. Joe Paradiso Responsive Environments Group, MIT Media Lab
http://www.media.mit.edu/resenv Seoul
August-09
Living with Ubiquitous Sensing and Dynamic Responsive Media
2
JAP 4/08
A Historical Perspective
3
JAP 4/08
…to experimental high-energy physics
PhD with Ulrich Becker and Sam Ting (MIT LNS) 1981
4
JAP 4/08
It’s great to do a thesis in Switzerland
CERN is a city of physicists Geneva is between Alps and Jura Any European culture is a short train ride
5
JAP 4/08
PreHistory - 1960
Me - 1976 1974
1976
At Ars Electronica Festival – Sept. 2-9, 2004
8
JAP 4/08
World’s largest home-designed-built synth (1975-1985)
Newton living room - 1986
Musical Interfaces
Sensor Networks as Extension of the Nervous System
Cast our awareness across space, time, scale, modality…
11
JAP 4/08
Bootstrapping a Ubiquitous Sensor Infrastructure
• Sensor networks are the foot soldiers at the front lines of ubiquitous computing
• At this point, few if any customers will buy an ensemble of “UbiComp” sensors
• They will aggregate from established devices – Home security, appliances, utility devices, entertainment…
Just as the web sprouted from a networked ensemble of personal computers, true “ubicomp” will arise from an
armada of networked devices installed for other purposes.
12
JAP 4/08
Power Strips are Everywhere
• Needed in Homes, offices, especially the Media Lab! • Sensors are becoming commodity items
– Cost of adding sensors to a design is becoming incremental • Power strips are ideal base platforms for hosting a
sensor network – Ready access to power – Power line can be a network port – Can monitor the status of devices that are plugged in
13
JAP 4/08 PlugPoint – Power Strips as the backbone of a UbiComp Sensor Infrastructure J. Lifton, M. Feldmeier, Y. Ono (Ricoh)
Power Line provides energy & comm
Monitor current profiles, Switch individual sockets Hosts basic sensors (mic,
light, motion) Expansion Port for others Hub for wireless sensor
network
Collaboration with Ricoh Research
Army of Plugs
35 ON MEDIA LAB THIRD FLOOR
MIT Media Lab
Joe Paradiso / ResEnv
Distributed Acoustic Conversation Shielding
Yasuhiro Ono
Ricoh ML affiliate
Paper in ACM SANET 2007
17
JAP 4/08
Cross Reality
IEEE Pervasive Computing July 09 MIT ML, MIT CMS, Stanford, ETH, University of South Australia, Georgia Tech, Sun, etc.
Shadow Lab - Binding real sensor data to virtual worlds
Simple sensor apparitions to explore basic ideas - Energy use ➔ Amount of smoke
- Sound Levels ➔ Ripples - PIR Motion ➔ Waving Fronds - Temperature ➔ Frond Color
- Light Levels ➔ Frond Height
Third floor of ML built in Second Life
ResEnv Lab rendered in detail - other areas currently
derived from map
Sensor data piped in and interpreted as real-time
graphic phenomena
Lifton 07
Portal Details – Camera System Deployed to cover entire
building (~50 nodes) 3MP Camera Motorized Panning and Focus Dedicated Video DSP/ARM - (TI DaVinci chip) Real-time Linux OS LCD display (Touchscreen) Contains Spinner Gateway/
Sensor board (detailed on next slide)
45 distributed across Media Lab since October 08
Mat Laibowitz
Device Details – Spinner Gateway Works with or without camera
board Communicates with wearable/
mobile devices in mesh network Serves as reference beacon for
location system Ethernet or slave to linux board Audio system with DSP
AVR32 Environmental Sensing
Motion Humidity and Temperature Light Infrared Communication and
Detection/Proximity Talks to badge systems
45 all over the Media Lab
Power Switch
Invitation to Interaction
Live Portal-Portal Video Feed
Portals and Cross Reality
Real World
Virtual World
Will the market demand privacy?
People are bringing these sensors into their homes and lives on the backs of dedicated appliances and devices
If they don’t feel in control of them, they won’t buy them!
How will they manage their dynamic privacy in a world full of potentially invasive sensors?
- You can’t turn them all off because there are too many, and they are often attached to things that need to stay on or can’t be turned off…
world
group
user Accessible for user’s social network
Accessible for public
Accessible for data owner & close colleagues only
Video, Images, Clean Audio, fine localization
Scrubbed images, garbled audio, coarse localization
Image & audio features only, building-level localization
No ID, building-level features
User-Configurable Dynamic Privacy Settings
Privacy can depend on physical location, time, and inquiry identity
Nan-Wei Gong
NoNo Badge for Dynamic Privacy
Dynamic Privacy…
NoNo Badge for Dynamic Privacy
Dynamic Privacy…
Nan-Wei Gong
MIT Media Lab
Joe Paradiso / ResEnv
Privacy requests (button pushes) per user vs time of day
31
JAP 4/08
32
JAP 4/08
Public Misinformation…
“Human” Energy harvesting will do little for sustainability It will only be valuable in extending/eliminating batteries in
portable devices, wearable sensors, etc.
33
JAP 4/08
Sensor networks for energy conservation
• Leveraging dense sensor networks for optimal energy management – 40% of US energy is spent in buildings – Pervasive sensor/actuator network can reduce this
• Optimize heating, AC, lighting for Person not room • Anticipating behavior & build usage models over time
Wireless Monitor
Wearable Integrated vibration, T & H, Light @ µW Mark Feldmeier
Ceiling PIR 802.15.4 bridge
Ceiling HVAC Damper Control
Measure temperature, humidity, airflow, PIR motion Continuous Control of Damper
Open/Close Windows
37
JAP 4/08
Learning Comfort
38
JAP 4/08
Very Early Data – Learned Setbacks
39
JAP 4/08
SportSemble 3x HiG Accl, 3x LowG Accl, 3x HiRateGyro, 3x magnetometer
40
JAP 4/08 Network In Action
To be presented at ISWC 09
Return to Spring Training, 3/09
New device has onboard removable flash for continual data storage (RF synch), Bi-range full IMU
Soon measuring entire team and streaming data to MIT/MGH
Red Sox Spring Training - March 09
Video by Alex Reben
44
JAP 4/08
Exchange Business Card, Bookmark Demos
45
JAP 4/08
Bookmark data posted on website
46
JAP 4/08
Bookmark data posted on website
47
JAP 4/08
Badge Accelerometer & Audio Data
48
JAP 4/08
Interest Meter & Group Dynamics
Collaboration with Human Dynamics Group
Badge-Badge Badge-Demo
Uninterested
Interested
Uninterested
Interested
Classification Value = f (voice, motion, face-face time…)
80% Accuracy
49
JAP 4/08
Affiliated Wearers
Affiliated people tend to spend more time face-face and move together!
Accuracy of inferred affiliation: 93%
Face-Face Time (IR) Correlated Motion (accels)
50
JAP 4/08
Affiliated Wearers from Energy Only
Timekeeping for Talks
T – 3 mins T – 2 mins T – 1.5 mins T – 1 mins T – 30 sec
Out of time!! • 24 talks in the morning (research updates) • 5 Minute time limit on each! • Audience badges flashed time queues • We didn’t run over (first time ever…)!!
SPINNER • Sensate Pervasive Imaging Network for Narrative
Extraction from Reality
• Unites wearable human sensing with video capture • Use of wearable sensing allows access to subject/data
channels far beyond what can be achieved with standard image pixel processing
Mat Laibowitz - PhD finished fall 09
Query multimodal sensor networks in human terms (through narrative/stories)
Overall System Diagram
Behavior Models
Sensor /Action Models
Captured Events
(video clips w / sensor data )
Assembling of Narrative(video )
Meaningful Narrative Discourse
(video)
StoryModel
Design /Input
UserInterface
Sensor NetworkInternal Mappings
Sensor /Imaging Physical Network
Sensor Data
TrainingDataFE
EDB
ACK
FEED
BAC
K
Camera System
The Age of Ubiquitous Media
• Is here, whether you like or not
• Can record your entire life, too much video to ever watch
• Need a mechanism to create/extract meaningful content
Device Details – Spinner Social Sensor • Wearable on collar or as pin/
badge • Audio system with DSP for
analytics and CD quality recording
• Mini-SD Slot for copious memory • Compass for orientation • 3-axis Accelerometer & 2-Axis
Gyro • IR communication and line of
sight detection/proximity • 802.15.4 Radio with RF Location
engine • Captures social signal and group
dynamics • New Badge has OLED Display
System talks to Portals • Cameras will record whenever Spinner badges
are nearby • May also use wearable cameras (e.g., N95) for
1’st person perspective
First ever video edited automatically using sensor data and filtered
from a >40 node distributed camera network
Bright and calm - Active - Dark and calm
Story Arc…
59
JAP 4/08
Conclusions • Sensing, computation, and communication become tightly integrated and
commonly embedded • Low power and energy scavenging enable active nodes to be embedded and
“forgotten” • The Ubicomp infrastructure permiates our cities, our dwellings, our objects, our
clothing, and eventually our bodies – Pervasive-Wearable-Implantable
• The 5 human senses locked into our body are augmented by interfaces into ubiquitous sensor network data – Marshall McLuhan for real – Interface devices now - implantables some day – Omniscience…
• This infrastructure mediates everything – Collaboration, business, social interaction, resource use… – Context engines filter, represent, and manifest information
• Google for reality • Brave New World
– Privacy, security, promises vs. perils…