Apple Watch side App
Standalone App for data collection that connect with the iPhone App with the intention to increase the level of accuracy and bring new information to expend actual functionalities.
Apple Watch placement
We think that the best placement is on the ski on a mount. The exact placement has to be define and we will have to investigate the impact on the data collection too. We have to check if we will have to fool the skin detection security feature of the Apple Watch for the App to work properly.
The watch App should be quite straight forward. Collect data of the different sensors and send it to the main App for data interpretation, analysis and rendering.
Establish connection between the Watch and the Phone
The discipline should be chosen on the Phone App and based on that the phone will pass the info to the watch over Bluetooth so the Watch App will know what to record.
The Watch interface should :
only allow to establish a link/connection to the Phone App
Have a Start Capture bouton to set the Watch in a ready to capture state
Have a Stop Capture bouton to finish the data capture session.
Technical aspect that can influence the behaviour/workflow
See if we have to send the collected data at the end of the ski set
or if we can send data to the boat (and the coach for example) between each pass. We have to check if we can have a live link from the Watch to the Phone App while skiing too.
Limited intelligence could be built in to detect different state and trigger [login to view URL] exemple it doesn’t worth collecting data if the skier is in the water (not moving). Based on skier profile, data could be collected when the skier reach a certain speed and stop capturing data when speed drop bellow a certain value.
What we would like to know with the Apple Watch. Check if it is technically possible
Synchronise the data capture with the video recording of the phone. We should be able to associate a certain video frame with the Watch data capture at the same time
Speed of the watch
Direction of the watch
Tilt of the watch:
Front back (to determine how flat the ski ride on the water and the changing state)
Side to side or edge to edge (to determine the lean of the ski)
Recreate the path of the watch on the phone
Calculate the distance between the Apple Watch and the phone. How accurate can it be? Is it then possible to have a precise length of the rope? Or how can we solve this
In slalom, detect the passes and provide an accurate classic score (4@32 36mph)
In jump, calculate jump distance
Tricks will be address in a future as it will have to involve AI/machine learning to recognise data pattern
After we have a clear understanding on what can be done on the data collection side, the accuracy and the cleaning of the data (to eliminate noise) we will be able to address what data we want to show the user and how