DMP: Classification of Trick Shots in Rocket League Using Machine Learning
Year: 2025
Type: article
Abstract: This project applies machine learning techniques to predict the type of skillshot executed just before a goal is scored in the video game "Rocket League".Using data from the last 30 frames before each goal, a variety of player and ball features (position, speed, inputs, etc.) were collected and analyzed.The goal is to automatically classify the executed skillshot into one of seven categories. Originally developed as an exercise from a machine learning course, the project was expanded to apply FAIR data principles:- Data is stored in a university-internal DBRepo instance with a DOI.- Results and models are uploaded to the TU Wien Research Data platform (TUWRD).The source code can be found on GitHub: The source data is stored on DBRepo: 10.82556/a3cq-wb28 or direct link The output is stored on TUWRD: 10.70124/djrhq-kxg26 or direct link
Author: Pelka, Thimon
Institution: TU Wien
Language: English
Topic: Digital Games and Media
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W6893328825
FWCI:
Cites:
Cited by:
Related to: