Design and Implementation of a Computer Vision Systemfor an Autonomous Chess-Playing Robot

• Guillermo Larregay ^[1] ; Federico Pinna ^[1] ; Luis Avila ^[1] ; Daniel Morán ^[1]
  1. [1] Universidad Nacional de San Luis

     Universidad Nacional de San Luis

     Argentina

     
• Localización: Journal of Computer Science and Technology, ISSN-e 1666-6038, Vol. 18, Nº. 1, 2018, págs. 1-11
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • This work describes a mechatronic system composedby a robot arm that can play chess autonomously. Thesystem is based on an industrial-grade robot manipu-lator, a computer vision system, and an open sourcechess engine. Classification algorithms were imple-mented in order to detect whether a given chessboardsquare is occupied, and in that case, if the piece isblack or white. Such algorithms were compared interms of their complexity of implementation, execu-tion time and accuracy of predictions. To achieve anuniform illumination of the chessboard, a theoreticalmodel of an LED illuminance curve was used to findthe best orientation for each diode using a genetic al-gorithm. Both the support base for the LEDs and thechess pieces were made using a 3D printer. This im-plementation demonstrates the capabilities of the pro-posed vision-based system, whose complexity can beincreased in the future for a number of applications