Coding Stars 2.0 - Space Apps Challenge

Coding Stars 2.0| Sleep Shift Scheduling Tool

Awards & Nominations

Coding Stars 2.0 has received the following awards and nominations. Way to go!

Local Peoples' Choice Winner

Sleep Shift Scheduling Tool

Sleep loss and fatigue may lead to reduced performance and an increased risk to safety during many activities, including spaceflight. Your challenge is to develop an operational sleep shift scheduling tool that provides autonomous customization of a schedule for sleep, exercise, and nutrition to manage fatigue.

Smart Sleep Helmet

Summary

Our project primarily aims to enhance the sleep cycles of astronauts by creating a smart sleep analysing helmet with an accompanying app. The helmet comprises of a VR headset, noise cancelling microphone and a head band embedded with many brain probes and ear temp sensors to maintain suitable body temp for a comforting sleep. This helmet is worn prior to sleeping and the brain probe sensors produce PSG signals which are processed by a CNN. The output will assess the sleep-cycle and eventually optimize personalized schedule for each astronaut in our customized mobile app . Moreover, the VR reduces the exposure to bright light and can also play calm videos or sounds to reduce stress levels.

How We Addressed This Challenge

The smart helmet can provide a holistic experience to the user. The VR set helps with the light intensity as it will trick the brain into a certain level of darkness mainly due to the high resolution captures which gives a realistic experience. Moreover, the ear temp sensor senses the body temperature and adjusts the environment accordingly as they tend to complain about feeling "too hot or too cold."

Each astronaut will have their own helmet and which they would wear before sleeping. The helmet is also linked to an app which has 6 features namely Sleep, Schedule, Nutrition, Exercise, Relaxation and Positivity. For instance, if an astronaut is about to sleep but he feels cold. In this case, the helmet will recommend to eat a snack and listen to music or do some certain exercise for a fixed duration because it will increase the heart rate thus, naturally feel warmer without wasting energy. Approaching a natural methodology is usually better because they wont have to experience any side effects especially when they are up in space with a weak immune system and body. A Machine Learning model that uses convolutional neural networks will be used to predict if the astronaut is in REM sleep and measure the time they stay in that state to assess their sleep quality. We'll be able to control the day's schedule such as their tasks, kind of exercise to do and the food to eat. This schedule is not fixed and changes real time. The Relaxation feature in the app will enable the VR to showcase pictures and videos of the environment they love and are used to such as their living room back home. It will also play calm and comfortable sounds for them to have a better sleeping experience and calm their minds. It helps them get immersed into the experience itself and feel that they are doing their jobs as astronauts in the same environment on earth around their loved ones. It is important that astronauts feel comfortable and are relaxed so that they can concentrate in important tasks assigned to them in the mission.

How We Developed This Project

According to a research conducted by Laura and her colleagues at NASA, there are numerous factors that can affect the sleep of astronauts such as exposure to light frequently, misalignment of circadian rhythm, motion sickness, loud noise (as high as 72 decibels) from space shuttle etc which makes them feel fatigued during the day decreasing the quality of their work. Sleep is vital for rejuvenating the body and it is advised to obtain eight hours of sound sleep. Unfortunately, they are able to sleep only for three hours or less which inspired us to work together to enhance their sleeping experience. We as a team did extensive research and thought about the issue in their perspective. Since the space shuttle is their work and rest environment for missions that may sometimes extend to many months, there is a need to develop a mechanism to make them feel home, reduce stress levels and make them forget their worries momentarily. Hence, we formulated the idea of a smart helmet with a VR set and noise cancelling headphone to block external sounds. Using high resolution Vuze VR camera , various favourite spots on earth like astronaut's home is shot, and is made available through our app created using Ionic4 customized for every astronaut. This may help reducing the feeling of home sickness and declutter their mind to be able to sleep well. In addition, we also wished to analyse the quality of sleep and the stages mainly REM or Non- REM with the help of brain probes and predict using a convolutional neural network to be able to adjust their schedule for the day in our app. The CNN is implemented using Python on Google Collab using Keras or Pytorch. To showcase our helmet design, it was 3d modelled in the software SolidWorks .

How We Used Space Agency Data in This Project

We researched about the space data but unfortunately nothing provided open data sets to be utilized for our ML model. However, we referred research papers by NASA investigators such as Charles Czeisler, Laura Barger, and Steve Lockley for sleep, Scott M. Smith and Sara Zwart for nutrition and Alan Hargens for health.