project
play

Project Selection RESULTS OF THE FIRST SCREENING Screening process - PowerPoint PPT Presentation

Project Selection RESULTS OF THE FIRST SCREENING Screening process Ranking by votes Filtering self-votes Interest with respect to course topics and AmI features Evaluation of possible issues General statistics 44 ideas


  1. Project Selection RESULTS OF THE FIRST SCREENING

  2. Screening process • Ranking by votes • Filtering self-votes • Interest with respect to course topics and AmI features • Evaluation of possible issues

  3. General statistics • 44 ideas • 14 groups (1 person re-assigned) • 1 person out of groups Interest # of ideas Medium-high 6 Medium 7 Medium-low 1 Low 11 No-interest 19

  4. Group statistics • Group creativity – NULL (3 selected projects) – TAP (3 selected projects) – SmartAttack (2 selected projects) – The Ten-It-Eg Team (2 selected projects) – AMITEAM (1 selected project) – SmartUp! (1 selected project) – PacMan (1 selected project) – PiSquared (1 selected project)

  5. Selected ideas Idea Proponent Assignee Smart Gardener NULL NULL Smart Raise Your Hand Smart Attack Smart Attack Thermostat Reimagined PiSquared PiSquared Smart Management Of Smart Attack - Computers in Laib Mood Control TAP - Smart Notifications TAP TAP Intercom 2.0 PacMan PacMan Never be locked outside NULL YAAIG your apartment Save Light & work The Ten-It-Eg Team The Ten-It-Eg Team happily

  6. Selected ideas Idea Proponent Assignee Healthy Life Control SmartUp! - Smart Comfortable The Ten-It-Eg Team Tiger Team Lighting for the user Smart Pet Feeder NULL GGHF Consumes monitor & TAP - optimizer Environmental Noise AMITEAM AMITEAM Context lighting Teachers PU:TI Smart Baby Monitor Teachers IMA Smart Butler Teachers SmartUp! Treasure Hunting Teachers iMe

  7. Context lighting Exploit colored ambient • • AmI Features: lights (e.g., provided by the sensitive, adaptive, HUE system) to provide informational awareness transparent, for users: design and implement light and color ubiquitous (partly) patterns for providing unobtrusive notification of contextual information. For example for incoming mails, messages, tweets, people at the door, unconfortable or unsafe home states, etc. Must tackle multi-user settings and/or presence.

  8. Smart Baby Monitor • Design a smart baby monitor • AmiFeature: which allows to constantly check a baby for situations where a sensitive, adaptive, parent intervention is needed. It should be room-independent transparent, partially (exploiting the home sensors) and may permit different actions ubiquitous depending on the current context, e.g., open a video channel towards the parents (by selecting the most suitable device), open an audio channel for baby reassurance, re-direct the audio channel over the most suitable device, take autonomous actions such as playing relaxing music, shaking the baby cradle, use lighting for avoiding darkness fear, etc.

  9. Smart Butler • Vocal and/or chat-based smart • AmI Features: all butler which enables home inhabitants to delegate duties to • Min. Group size: 3 the home itself depending on current context, location, or get information like a human butler would do. For example: – Location-based or time-based reminders Scheduling of activations – Location-aware activation – Remote control – Task support (e.g., reading recipes – while coooking, habits reminder, etc.) – Information – Activity delegation(e.g. I’m only available for phone calling from parents)

  10. Treasure Hunting The ambient «intelligence» sets a • • AmI Feature: almost treasure to find and a set of riddles or puzzles, involving orienteering, art, all problem-solving, etc., which uncover the next step towards the treasure. • Can be deployed both indoor and • Min. Group size: 3 outdoor (indoor is easier for the final evaluation) • Exploits sensors (e.g., presence, consumption, button pressure, gps, etc.), actuators (e.g. Lights, voice, messages) and end-user devices to detect the players location and to deliver the next step • Must be multiplayer • Paths towards treasure might be changed at runtime by the system according to the current game context

  11. Next steps • Provide a complete system summary – As GitHub pages in the project repository • provided by the end of the week • Start defining requirements and architecture – in the GitHub pages of the project repository • Next Lab for – working on / finalize the documents – technology / tool scouting

  12. Now • Group finalization • Project assignment

  13. Smart Gardener: the self-learning watering system (4) • Smart Gardener is a 2 non-self votes • project that would allow Smart Watering • you to enjoy plants while • Proposal authors' team: lifting you from the task NULL of watering them. It AmI feature: autonomous, • would also optimize sensitive, adaptive, resources by avoiding trasparent, ubiquitous watering if rainfall is Soil Sensors must be • forecast in the near built/bought by the group future or if the day will Interesting project, shall • not be particularly hot. be deployed in the Lab for Feedback from the evaluation plants allows to tailor Initial ranking: medium- • irrigation cycles to their high-interest specific needs without manual input.

  14. Smart raise your hand system (2) • 1 non-self votes • Problem • In most of the laboratory exercises all students call the professor (or his assistants) a lot of times to check • Student assistance their works or ask help. This is a waste of time because of the professor has to run around the lab often to improvement explain the same things; the student has to hold his hand raised waiting for the professor. • My Idea: • Proposal authors' • Every computer has a device to detect that the student needs help as if he has raised his hand. The team: Smart Attack student can either ask a generic help or specify witch point he needs help. The student is placed in a queue. Every professor has a device(ex. smart watch) that • AmI feature: almost all signals the next student to help and if a lot of requests are on the same point it suggests that he could explain it to all students on the blackboard. • Needs more details on Is it smart ? • Every time that a class ends the entire exercise the • the learning part and system collects statistics on requests of help. In the next exercise that involved the same class, a display on the device for will suggest in which part of the laboratory the student should sit. The purpose is to place the students which need more helps near. So the system detecting attention learns time by time how to is better place students. In this way the teacher have not to run from one side of requests the laboratory to the another. • The statistics can also be used to know which arguments are more critic and the professor can • Initial ranking: explain them again in classroom. medium-high-interest

Recommend


More recommend