Difference between revisions of "PRE2015 3 Groep4"
|Line 103:||Line 103:|
== To do ==
== To do ==
* alles doorlezen en spellingscontrole
* alles doorlezen en spellingscontrole
== Links ==
== Links ==
Revision as of 14:16, 20 April 2016
This is the wiki page of Group 4 of USE Project: Robots Everywhere.
For this course, we've made a prototype of a "Smart Bedroom" using an Arduino unit connected to a laptop. Our idea of a Smart Bedroom is a bedroom filled with electrical modules that are controlled by a central computer, which automatically commands them to simulate the sunrise and morning sounds in order to wake up the user in the most natural way possible. So the Smart Bedroom basically replaces your alarm clock and makes sure you're well rested and cheerful in the morning. The sunrise is simulated by a combination of a light sensor mounted on the outside wall of the bedroom and automatically controlled blinds and lights. Another module of the Smart Bedroom is a thermostat that gradually heats up the room when it's time to wake up the user. Morning sounds are provided by your smartphone, which also acts as a user interface for the entire system. Furthermore, the smartphone records the sounds the user makes when sleeping, most importantly his/her breathing, in order to determine how deep he/she sleeps, which is necessary to find out when it's a good time to wake him/her up. Through research we found that waking up during light sleep is much more comfortable than during deep or REM sleep. Setting the time period during which you want to wake up is also done via the smartphone.
For our prototype, a few concepts had to be simplified or even scrapped entirely. Naturally, the thermostat and blinds were off the table, which also eliminates the need of a light sensor. None of us have experience in making apps, so we decided not to use a smartphone as well. Instead, the laptop acts as a kind of user interface. The prototype consists of three scripts written in MatLab which are aided by an Arduino unit. The Arduino contains a microphone that records the user's breathing sounds and a LED that produces light. The laptop translates the sound measured by the Arduino into a graph showing the sleep deepness and uses that to decide when to start the wake up procedure, which consists of gradually turning on the LED and producing morning sounds with the laptops speaker. When the user has woken up, which he/she indicates by pressing any button on the keyboard, the opportunity is given to provide feedback about the sound volume and light brightness, which is taken into account the following nights.
- 0888465 - Thomas Bardoel
- 0942166 - Wesley van den Broek
- 0887435 - Jeroen Ermers
- 0888864 - Luuk Ladegaard
- 0887879 - Roel Montree
- 0835870 - Jeroen Verbakel
Our idea is the Smart Bedroom. This is a bedroom which will be controlled by a so-called smart alarm. The smart alarm will analyze the person in the room when he is awake and when he sleeps. The smart alarm can also control the lights and blinds in the bedroom. The purpose of a smart alarm is to let the person sleep as well as possible. This alarm can be very useful for people with insomnia.
The Smart Bedrom can help the person fall asleep better by helping the person to get asleep, waking the person after a full sleep cycle and waking the person with special light which will be experienced as a sunrise. The smart alarm has an integrated tool which can measure the heartbeat, which gives information about the sleeping process. The person can also give feedback. For example, he can rate his experience after every night, or he can only report the alarm if he experienced something unpleasant. The smart alarm will use this feedback the next nights to improve the sleep experience of the person.
The Smart Bedroom will be controllable on your smartphone. You can set the wake up time, or wake up interval, on your smartphone easily. The smartphone communicates with the software of the Smart Bedroom. This software controls the blinds, lights, sound and temperature in your room. When the person is waking up, he can give a rating on his smartphone. But he can also see his sleep pattern on his smartphone of all the previous nights.
A very precise method for measuring the heartbeat is with infrared radiation. But this method is still in development and will be very expensive. So we will use another method, namely sound recognition. This method measures the sound of the person breathingand translates this to a heartbeat. But a downside of this method is that it has to be calibrated over a few nights to get a good estimate. Sleeping together can also be a problem. But this method is able to be used nowadays and is cheaper than infrared radiation.
Here's a short story to illustrate our idea:
John is a student at the University of Technology Eindhoven. He has trouble waking up in the morning, so he decided to invest in a Smart Bedroom. When John goes to sleep, he sets the wake up time to anywhere between 7am and 7:30am on his phone. Now it is time to sleep. The bracelet John is wearing is measuring the heartbeat and sends data to his smartphone. Also, his phone checks his movements to see in what phase of the sleep cycle he is. Around 7am, the Smart Bedroom notices that John almost completed a sleep cycle. Since John will have completed this sleep cycle during the wake up interval, the Smart Bedroom lets him sleep a little longer. When John is close to finishing his sleep cycle, the Smart Bedroom instructs the windows to let more sunlight pass through and also raises the temperature. When getting closer to finishing the sleep cycle, more and more sunlight will pass through the window and the temperature will keep on raising. This allows a much more natural wake up than usual. The sleep cycle has ended but the smart bedroom notices that John is still asleep even though all sunlight is going through the window and the temperature is raised to the maximum level allowed, so it decides to play a relaxing sound which builds up in volume to force a wake up. Finally John wakes up at 7:20 am and he is not as sleepy as usual. He is able to get out of bed without any problems and is ready for his day at the university. John also rates his sleeping experience on his phone and the alarm will use this information for the next night.
The Smart Bedroom is a dream for many people. Here, the USE-aspects of the Smart Bedroom will be discussed.
The user will benefit from the smart bedroom because this integrates his blinds and lighting into his smartphone. This is the first step towards home automation. By crossing the barrier of combining your living space with your mobile devices, it becomes very easy to expand on this.
The Smart Bedroom helps to battle tiredness, and therefore also stress, which can also be considered a benefit for the society. Besides tackling tiredness in general, the Smart Bedroom is great for people with problems getting asleep.
Due to the integrated lighting and blinds in the bedroom, it becomes possible to sleep during the day. This possibly opens up all kinds of new lifestyles, such as working in multiple shifts during 24 hours and decreasing the morning and evening commute by spreading the starting hours throughout the day.
A good night's rest greatly improves productivity, which is good for both society and enterprise. Through the Smart Bedroom, we can open a whole new market in home automation, enabling many new jobs, startup companies and other benefits to enterprise. If you once again consider working night shifts, you'll see this is beneficial to enterprise as well, in decreasing the size of offices, as not everyone has to work there at once.
- The person who sleeps in the smart bedroom
- Other users sleeping in the smart bedroom
- People living with the primary user
- Scientists (data collection on sleep)
- Employers (more productive personnel)
- Engineers (that install and service the smart bedroom)
- Medical personnel (for people that need monitoring when sleeping)
Listed below are the needs of the primary, secondary and tertiary users. We will mainly focus on the needs colored in blue.
The person who sleeps in the smart bedroom
- The user should be able to set the time. The option to synchronize the time with the time according to your smartphone would be very helpful.
- The user should be able to see what time it is in the dark. The user could simply open his/her phone, or there could be an option to continuously display the time. This would, however, drain its power fairly quickly, so it's not recommended.
- The alarm clock should not have a snooze option. Being able to go back to sleep after the alarm has woken you up is proven to be bad for your health.
- The alarm should have a volume control. This could simply be the phone’s volume control.
- The user should have the possibility to choose between three options: no alarm, a standard alarm which wakes you up at a specific time and the smart alarm.
- In case of the standard alarm, the user should be able to choose when he/she wants to be woken up and what sound the alarm should make: a repeating pervasive tone, the radio or perhaps a song which is stored in the phone.
- In case of the smart alarm, the user should be able to indicate in which time frame he/she wants to wake up. If the system can’t find a suitable moment to wake up the user, the standard alarm should go off at the end of this time frame.
- Perhaps the user should be able to indicate when it’s a suitable moment to be woken up. However the user probably doesn’t know a lot about sleep patterns and such, so it’s not a necessary option.
- The user should be able to choose what the standard and smart alarm control, like the lights, the blinds and the temperature of the room. These could also have different settings depending on the user’s preferences.
- The system should be able to deliver understandable information to the user about how he/she slept that night, the nights before and in average. The user should be able to decide whether or not he/she wants the information to be (anonymously) shared on the internet for statistical purposes and/or be available for his/her doctor.
- The system shouldn’t badly influence the user’s sleep in any way.
- The system should be adaptable to a changing bedroom layout, which it probably is because most is done by the user’s smartphone, which is of course portable.
Other users sleeping in the smart bedroom
- Because the smart alarm wakes its user up based on the sounds the primary user makes during sleeping, it perhaps will not work when there are more people sleeping in the same room. In addition to that, in most cases there is no moment where it’s for everyone ideal to wake up, so using the smart alarm would be useless. The users could still use the standard alarm, of course.
People living with the primary user
- These people do not want to be involuntarily woken up by the alarm. For this reason, the sound the alarm makes should not be too loud to wake up the people living with the primary user. It should be noted here, that the alarm should still have the capacity to go very loud, as some people might require a louder alarm to wake up.
Scientists who collect the data on how people sleep
- The scientists should have a direct link to the database of shared information about people’s sleep.
- The information needs to be complete (so as much information as possible), structured and easily exportable to other statistical programs.
Employers of the users of the smart bedroom
- The employers benefit more from active and well-rested employees.
Engineers that install and service the smart bedroom
- The system should use mainstream components like normal screws, standard European cable inlets and outlets etc.
- When installing the system, the components should arrive in an orderly fashion, so similar parts are grouped together in a plastic bag for example, and a detailed instruction manual should be included. If necessary, the system should be delivered with a few tools which are specifically designed for installing it.
- The system should come with a detailed troubleshooting manual which suggest what could be wrong and how it could be fixed if there’s a problem with it.
- If necessary, the company behind the production of these smart bedrooms should hire and train personnel who are specifically employed to deal with this system.
Medical personnel for people who need monitoring when sleeping
- The medical personnel should have a direct link to the information of their patients.
- The information needs to be complete, structured and easily exportable to other statistical programs.
- In case the medical personnel has no experience with handling data, the information needs to be presented in a way which is easy to understand, with graphs for example. The presentation should still be able to show important aspects which are vital for the personnel’s diagnose of the patient’s sleep patterns.
- alles doorlezen en spellingscontrole
Below are several links which have each have another function. In the first link 'System input and output' a overview is shown of our dream and the product we realy gona made. In the link 'Measurement plan and experiments' is the measurements located. The results of the experiments we did, the made trade-offs and justification of the method of measuring can also be found at this link. In the next three links, 'input', 'algorithm' and 'output', the whole code is described. In the link 'input' is the arduino code described for sampeling the sound. In the link 'algorithm' are the matlab scripts explained which finaly determined the ideal moment of waking up. And the link 'output' described the working of the output script and the made trade-offs. In the 'results' is a brief overview of our final product. Also all actions to start the Smart alarm are located there. Finally, in the link 'conclusion', not only the conclusion will be given, but also the futher improvements and the reflection.
In the link 'literature' can be found some background information about some specific topics. For example about the sleepcycle in genaral, the ideal moment of waking up and the influence of light, temperature and sound during your sleep. The information about this topics is used by making decisions during the construction of the algorithm.
- ↑ "Lecturing: Snoozing and Losing", S. Browning and K. Divoll, 2010, http://digitalcommons.georgiasouthern.edu/sotlcommons/SoTL/2010/81/