Skip to main content

RED.ES

Funded by the European Union – NextGenerationEU, Government of Spain, Red.es and the Recovery, Transformation and Resilience Plan

The project with file number 2021/C005/00149717 and title “Algorithmic engine for training Esports skills” has been a beneficiary of the “2021 call for grants for Research and Development Projects in Artificial Intelligence and its integration into value chains, C005/21-ED”.

The main goal of the project is to develop a technology based on AI algorithms and a training methodology aimed at increasing players' performance, as well as reducing the physical, psychological, postural, nutritional and visual pathologies associated with practice. The idea is to bring a sports approach to Esports players by developing software that lets them train the cognitive activities they need. Specifically, in this project the company focuses its efforts on the following specific objectives:

  • Development of a data mining and processing layer to extract and analyse data from LoL matches, drawing on published databases of professional leagues and on the training matches accumulated at United Gamers.
  • Development of algorithms for modelling player behaviour based on their decisions in training sessions and previous matches. Using the information from the data mining and processing layer, the aim is to create an algorithm that reinforces players' training based on the weaknesses detected. It will also include the development of ML algorithms to learn a player's preferences from the decisions they make and thus predict which skill or skills they should train.
  • Development of adaptive algorithms that adjust training difficulty according to the player's real-time actions in each match and their previous modelling, in order to dynamically create fully personalised training for our users.
  • Development of intelligent attack bots whose movement and behaviour are defined by AI algorithms. This objective involves researching techniques such as reinforcement learning and a supervised long short-term memory network (LSTM) to create a LoL bot able to simulate plays and attack strategies for training sessions. The aim is to bring AI to video game training for the first time, through automatic attack bots that respond to the user's actions.

Funded by the European Union – NextGenerationEU.