Since automation of tasks with wide variance of position and shape such as bin-picking, finishing and assembling are demanded, more and more robots are becoming adaptive with force-torque sensor and vision sensors. PC-based integration is one of the best solution to construct such sensor-based robot systems because of flexibility and extendability. The robot system become more complex, the number of tuning parameters increases. To solve this problem, we propose a new method to reshape the response characteristic of industrial robot using external controller for reducing the tuning control parameters. In this method, dynamics of a robot arm is measured via real-time communication, its mathematical model is estimated, and then the controller of an external PC is designed. In this paper, the proposed method is validated through simulation and experiments.

In this paper, we develop a waterproof suit for life-sized humanoid robots and propose a walking control method considering reaction forces from water for extending the region where the robots can move. The robots that wear a removable waterproof suit are easier to maintain and more versatile than the robots whose each component is manufactured to be waterproof. Unlike on land, humanoid robots are influenced mainly by the two forces due to water: buoyancy and drag. We estimate buoyancy online using force sensors at the ankles and consider the estimated buoyancy when generating a walking motion. The robot modifies footsteps in order to deal with large disturbances including drag force. We verify the effectiveness of the proposed methods through an experiment in which a life-sized humanoid robot walks down and up stairs, and walks fast in water.