Motion Control of Humanoid Robots from Human Motion Database

2017 Fiscal Year Final Research Report

• Project/Area Number: 26700021
• Research Category: Grant-in-Aid for Young Scientists (A)
• Allocation Type: Partial Multi-year Fund
• Research Field: Intelligent robotics
• Research Institution: Osaka University (2017); The University of Tokyo (2014-2016)
• Principal Investigator: TAKANO WATARU 大阪大学, 数理・データ科学教育研究センター, 特任教授(常勤) (30512090)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: 知能ロボット / 運動計画 / 運動生成 / 統計モデル

Outline of Final Research Achievements

Human whole body motions are encoded into a stochastic model, which allows for a humanoid robot to generate a human-like motion. It requires to generate the robot motion satisfying two conditions. One condition is maintaining a profile of training motions. It implies that the humanoid robot should behave as humans. Another is adapting to the external environment. The environment is varied, and the current environment should be different from that during training. The robot must perform adaptive motion to a new situation. These two conditions are measured as the likelihood of the motion generated by the stochastic model, and distance from the desired motion dependent on the environment. The combination of these two measurements designs the objective function. This project has developed an algorithm of searching for an optimal motion that maximizes the object function. This research provides the fundamental technologies that are necessary to integrate humanoid robots into our daily life.

Free Research Field

ロボティクス