meta data for this page
This is an old revision of the document!
DEVS
Use this template to document your progress! (Add notes anywhere you deem necessary)
1. Group Introduction
- Dũng Nguyễn
- Eric Rode
- Victor Olasehinde
- Samad Abdus
2. Initial brainstorming ideas/concepts
- Using temperature sensor to nudge users not to use air conditioner when it is unnecessary or the air conditioner temperature is much lower than outside temperature. System can automatically opens the window, activate the fans…
- Gas sensor to prevent gas leakage, system can turn on the fan and open the window for ventilation, show warning light to users…
- RFID sensor to detect entrance and give guidance to users (wash your hands…), raise warnings if users don't follow
- Steam detector after showering to do ventilation
- PIR sensor to nudge people to exercise between the day
Smart bedroom: - Sense temperature, humidity while sleeping to adjust for better sleep - RFID to make user get rid of their phones before sleeping - Light changing while sleeping - Open the window in the morning - Play white noise while sleeping or alarm in the morning - Gas monitor to detect gas leakage while sleeping
3. Day 2 Presentation slides
4. Finalized Idea, description & Functions
Proposed main features (Use Cases), sorted by priority:
- Prevent phone usage
- Coffee-drinking habit alert
- Turn off unnecessary appliances
- Sleeping time reminder
- Automatic exposure to nature light in the morning
- Sleeping environment optimization (temperature, light, ventilation…)
- Noise and activity detection at night
Business logic:
- Prevent phone usage: If the phone is not present in the RFID sensor after the sleeping time, play warning noise and message
- Coffee-drinking habit alert: If user push the coffee maker button (left) after 5 PM, show message and alert to ask if the user is sure. If he presses again in the next 10s then a coffee is still brewed, after 6 PM refuse to brew coffee, otherwise proceed to emulate brewing a coffee
- Turn off unnecessary appliances: Once it is past bedtime, display a message to send the user to bed and do red LED light and buzzer. Once the phone is detected (relate to Prevent phone usage), stop sending user to bed and turn off all appliances. Send the system to night mode (window and door get looked, lights fan and coffee machine turn off)
- Sleeping time reminder: One hour before sleeping time show a reminder on the lcd and start dimming the light every 10 minutes. At sleeping time it should stay on until Prevent phone usage turns it off.
- Automatic exposure to nature light in the morning: One hour before Wake up time open the window and turn on the light slowly (reversed to sleeping time reminder). Once its wakeup time end night mode.
- Sleeping environment optimization (temperature, light, ventilation…): During night mode regulate the temperature in the room. The optimal sleeping temperature should be configurable (standard 17 degree). If it is hotter start the fan and if the humdity is above 60% open the window. If below these values close the window/stop the fan.
- Noise and activity detection at night: If the PIR detector detects movement in night mode, then play a song with buzzer so that the person relaxes and falls asleep again.
5. Future Improvements
- Automatically adjust the sleeping schedule of user by having them pushing the button when they wake up
6. SusAF Analysis
The Smart Sleep Assistant has 7 main features based on the research using SusAF. Possible effects found are presented for these features considering social, individual, environmental, economic and technical dimensions. A few possible actions were identified and are presented at the end of each feature.
1. Prevent Phone Usage Effects:
Improved sleep hygiene (Individual)
Reduction of digital addiction (Social)
Potential resistance from users (Individual - Negative)
Actions:
Gradual enforcement; use messages before activating warning sounds.
Provide override options for emergencies.
2. Coffee-Drinking Habit Alert Effects:
Reduced caffeine-related sleep disturbances (Individual)
Increased user awareness of consumption habits (Individual)
Potential frustration or non-compliance (Social - Negative)
Actions:
Provide a bypass with confirmation.
Collect data to optimize timing alerts over time.
3. Turn Off Unnecessary Appliances Effects:
Energy conservation (Environmental)
Safety improvement (Social)
Reduced electricity costs (Economic)
Actions:
Smart schedule optimization based on actual usage patterns.
4. Sleeping Time Reminder Effects:
Encourages regular sleep patterns (Individual)
Reduces screen time exposure at night (Social/Individual)
Potentially intrusive if not customizable (Negative)
Actions:
Make reminders adaptive and user-configurable.
5. Automatic Exposure to Natural Light Effects:
Aligns circadian rhythm with daylight (Individual)
Enhances mood and productivity in the morning (Social)
Reduces need for artificial lighting (Environmental)
Actions:
Adjust based on sunrise time dynamically.
6. Sleeping Environment Optimization Effects:
Improved sleep quality through ideal room conditions (Individual)
Energy-efficient ventilation and fan use (Environmental/Economic)
Actions:
Allow customization of temperature and humidity thresholds.
7. Noise and Activity Detection at Night Effects:
Enhances feeling of safety and calm (Social/Individual)
May cause disturbance if false positive (Technical/Negative)
Actions:
Use multiple sensors to validate activity before reacting.
LIST OF EFFECTS
SOCIAL DIMENSION
High likelihood - High impact
Positive:
Enforced digital detox improves quality of interactions.
Safer environment by shutting off appliances.
Negative:
Resistance to imposed behaviors (e.g., phone removal, coffee restrictions).
INDIVIDUAL DIMENSION High likelihood - High impact
Positive:
Better sleep from controlled environment and reduced caffeine.
More consistent circadian rhythms.
Negative:
Discomfort due to automation override (e.g., blocked coffee brewing).
ENVIRONMENTAL DIMENSION High likelihood - High impact
Positive:
Lower electricity use due to appliance management.
Natural light use reduces carbon footprint.
ECONOMIC DIMENSION High likelihood - Low impact
Positive:
Marginal cost savings from reduced energy use.
Fewer replacements due to appliance overuse.
TECHNICAL DIMENSION Low likelihood - High impact
Negative:
Risk of system failure or false positives causing sleep disruption.
Hardware dependency (sensors, window motors) increases maintenance need.
7. Behavioral Change Analysis
Feature | Intended Behaviour Change | Data to Track | Device/Input Source |
Prevent phone usage | Reduce phone usage before bed | Phone presence on RFID; screen time after bedtime | RFID sensor, phone screen-time tracking |
Coffee-drinking habit alert | Reduce caffeine intake in evening | Time of coffee button press | Button input with timestamp |
Turn off unnecessary appliances | Reduce distractions, energy use | Appliance state after bedtime | Smart switches, time logs |
Sleeping time reminder | Encourage regular bedtime | Adherence to bedtime schedule | System clock, user activity (PIR, phone use) |
Natural light exposure in morning | Align circadian rhythm | Wake-up time consistency | Time of PIR activity or wearable wake detection |
Sleep environment optimization | Improve sleep quality | Temp/humidity levels, fan/window state | DHT sensor, actuator logs |
Noise & activity detection | Detect restlessness | Motion events at night | PIR sensor |
Self-reported sleep quality | Track subjective improvements | Daily score (1–10 scale) | Manual user input on screen or app |
Metric | How to Measure | Expected Change |
——————————— | ———————————- | ————————— |
Average bedtime | Time of last motion/phone use | Earlier and more consistent |
Wake-up time | First PIR movement or wearable log | More consistent, earlier |
Phone usage after bedtime | Screen time log, RFID absence | Decrease over time |
Coffee intake after 5 PM | Button press logs | Fewer presses after 5 PM |
Room conditions | Temp/humidity logs | More time in optimal range |
Nighttime disturbances | PIR sensor logs | Fewer movements at night |
Sleep score | User self-report & wearable data | Higher over time |
Sleep stages (REM/deep/light) | Wearable logs (e.g., Oura) | Higher % of deep/REM sleep |
Data Collection & Tools Sensors and Devices:
- RFID sensor – Phone detection
- PIR motion sensor – Movement before and during sleep
- Button input – Coffee habit monitoring
- Temperature & Humidity sensor (DHT11/DHT22) – Room environment
- Smart appliances – Logs of on/off state
- LCD Screen or App UI – Sleep reminders, feedback, manual input
- Wearables (Oura Ring, Fitbit, etc.) – Sleep stage and heart rate data
Storage:
- SD card or cloud logging (e.g., Firebase, Google Sheets)
- Time-stamped logs for sensor data
- Secure, anonymized user logs
A. Quantitative Analysis: Track changes week-over-week or month-over-month
Key comparisons:
- Average bedtime before/after intervention
- Caffeine intake frequency at night
- Sleep duration and stage improvement
- Night disturbances (motion events) reduced
- Use graphs: sleep score trends, usage heatmaps, condition logs
B. Qualitative Analysis:
- Weekly reflections: user rates sleep quality & experience
- Note perceived stress, restfulness, energy levels
- Interview or survey users to understand comfort, usability