Baseball pitchers use a variety of techniques to deceive the hitter, including varying the location, speed, and spin of the pitch. Several pitch types are often talked about in baseball lingo: 4-seam and 2-seam fastball, curveball, sinker, slider, changeup, knuckleball, etc.
Trackman has stadium installations in all Major League Baseball stadiums and in most Minor League stadiums, tracking every pitch thrown during games. Taggers classify this vast number of pitches during the game into the pitch categories mentioned above.
The Trackman units measure all relevant parameters of the ball flight using Doppler radar and computer vision tracking techniques.
The aim of the project is to use machine learning to classify pitches from the measurement data provided by the Trackman unit.
DTU supervisor could be a DTU Space employee.
I samarbejde medTrackman / Thomas Sørensen, email@example.com
Knowledge of machine learning techniques. It's an advantage to have some knowledge of – or interest in – fluid dynamics and how spin can influence the drag and lift.