Volley

volley

Will You Face Superhuman Opponents at the Next Olympics?

Ukraine's Rapid Push to Deploy AI-Enabled Drones for Battlefield Supremacy

This Cop Sabotaged His Own Cases to Undermine a Prosecutor: Why He Did So

Startups of The Year 2023 Winners: North America 

ROGER HAMLEY'S CONFESSION.

The Last Will of the Old Squire

U.S. Supreme Court Justice Clarence Thomas and His Powerful Billionaire Friends

মার্কিন সুপ্রিম কোর্টের বিচারপতি ক্লারেন্স থমাস এবং তার শক্তিশালী বিলিয়নেয়ার বন্ধুরা

Clarence Thomas, Richter am Obersten Gerichtshof der USA, und seine mächtigen milliardenschweren Freunde

El juez de la Corte Suprema de Estados Unidos, Clarence Thomas, y sus poderosos amigos multimillonarios

Clarence Thomas, juge de la Cour suprême des États-Unis, et ses puissants amis milliardaires

अमेरिकी सुप्रीम कोर्ट के न्यायाधीश क्लेरेंस थॉमस और उनके शक्तिशाली अरबपति मित्र

米国最高裁判所判事クラレンス・トーマスと彼の強力な億万長者の友人たち

미국 대법원 판사 클라렌스 토마스(Clarence Thomas)와 그의 막강한 억만장자 친구들

O juiz da Suprema Corte dos EUA, Clarence Thomas, e seus poderosos amigos bilionários

Судья Верховного суда США Кларенс Томас и его влиятельные друзья-миллиардеры

ABD Yüksek Mahkemesi Yargıcı Clarence Thomas ve Güçlü Milyarder Arkadaşları

Thẩm phán Tòa án Tối cao Hoa Kỳ Clarence Thomas và những người bạn tỷ phú đầy quyền lực của ông

美国最高法院法官克拉伦斯·托马斯和他有权势的亿万富翁朋友

Part 2: Real Motion Data from Apple Watch and Quick Analysis In the , I demonstrated how high-frequency motion data can be collected during a tennis session using my sample app. In this part, I'll provide a sneak peek into the data and try to extract some high-level learnings from it. previous part How I Collected the Data I play tennis once a week, so once I made , I was able to quickly take it to the real world to obtain the sensor data. I'm right-handed, so for better accuracy, I wore the watch on the right wrist even though I usually wear it on the left. As my simple iPhone sample app only keeps the data in memory and there should be an established connection between the counterpart apps, I had to keep my phone in my pocket. the app Only during the session, I realized that the iPhone app has to be in the foreground, otherwise the messages are not delivered. I made several recordings, each of them represented a single rally. On average, the warm-up rallies were 6-8-shot long. WatchConnectivity Accelerometer Data For a typical rally, its accelerometer file would be about 1–1.5 Mb. from that warm-up session that represents roughly a 19-second series with 15651 readings – remember, it's an 800 Hz sensor! Here is a sample file Here is a simple of the acceleration measured in Gs across all three axes: line-chart visualization We can clearly see the spikes just before and after the shot, with the Z-axis showing the highest amplitude. The Z-axis is orthogonal to the watch screen surface. That's why it's no surprise that it gets the largest acceleration when the wrist moves toward the ball. I just didn't expect that the racquet acceleration could reach 30G! Y is pointing from the bottom to the top edge of the watch, and X is basically the continuation of the digital crown. That's why it shows the least amount of movement. A single-shot close-up is also interesting (how acceleration is thrown back and forth over a tiny period of time): Rotation Data is similar to the accelerometer one, but here, it's radians per second for each axis. It shows similar peaks around the same timestamps; however, its readings are 4x less frequent: The gyroscope data Interestingly, the X axis shows higher amplitude than Z here, and each axis has a distinct pattern. As X is the digital crown axis, higher values here may mean more spin that comes from the wrist. Conclusion and What's Next This was just my first tennis session with this app, and my main goal was to see if it actually worked as expected and if the data had some actual meaning, even if presented as a simple line chart. Now, I clearly see a correlation between the line and its corresponding shots – for example, higher amplitude seems to mean a higher ball speed. I plan to eventually extract more useful information from the raw data – such as shot frequency, time to contact, distinguishing backhand vs forehand vs volley, approximate ball speed, and so on. To start, I could try to apply a median filter and then find local peaks. In addition, here are the ideas for improvements that I'm going to make to the app to be able to better analyze the data in future sessions (I am not sure if all of them are feasible and quick to build, but they should definitely help): Persist data onto disk on the watch during the session and transmit it as a large file – to avoid having to keep the iPhone app nearby and active. Preview the line charts right on the iPhone. Put the accelerometer and the gyroscope data onto the same timeline. Add a way to map the data recording to the actual rally – maybe by shooting the video of the practice on the counterpart iPhone and then synchronizing it with the motion data. I'm curious about what you think can be done with this raw data. Please do share your suggestions in the comments!

Getting High-Frequency Tennis Motion Data from Apple Watch [Part 2]

Consideration of Early Hypotheses

THE DUTY OF AMERICAN WOMEN TO THEIR COUNTRY.

Stories, News & Mentions about VOLLEY

Volley’s Stories on HackerNoon