Development of a Generalized Whole Body Motion Architecture for a Humanoid Robot with Applications in Soccer Robotics

Abstract

The field of humanoid soccer has become quite popular over the past decade, which can be attributed to the annual international robotics competition RoboCup. The competition presents new challenges every year, solutions to which are provided by the participating research teams, resulting in tremendous amount of qualitative research development in the field of humanoid robotics. One of the major leagues of the competition is the Standard Platform League (SPL) which comprises of a soccer game between Aldebaran Nao humanoid robots and has given rise to the development of numerous motion planning techniques for solving highly complicated, dynamic soccer related tasks. The aim of this study is to develop a generalized whole-body motion architecture for humanoid robots that can be used to design any dynamic motion task in real-time based on the required input optimization parameters. Furthermore, to demonstrate the capability of the architecture, the three most important tasks of the Soccer gameplay, that is, a dynamically stable omni-directional kicking engine, omni-directional walking engine, and a optimized targeted diving motion will be developed. The study will be carried out as a part of Team-NUST, the official RoboCup SPL team from Pakistan, and will result in significant improvement in the performance of the team for the upcoming competitions.