High-Accuracy Driving Control of a Stone-Throwing Mobile Robot for Curling

Abstract

In this study, a novel mobile robot that can throw precisely a stone in a curling match is introduced. This mobile robot plays alongside a human team. Several requirements for developing the mobile robot are defined so that it follows the game rules and operates on an ice field. A static force analysis considering the operating conditions of the mobile robot is conducted to derive the required motor torque and the optimal mechanical design. For a successful stone-throwing, the mobile robot must present a high driving performance on an ice surface with a target velocity and an angle provided by an autonomous strategy generator. To achieve the target velocity on an icy surface, a velocity control algorithm based on slip ratio control (SRC) is proposed. Moreover, to precisely achieve heading angle control, a heading angle controller including a steering angle disturbance observer (SA-DOB) is introduced. The mathematical model of the SA-DOB is derived by a dynamic analysis of the mobile robot. Several experimental results verify the proposed motion control algorithm and the feasibility of the mobile robot as a curling robot to compete with a human team. This study is motivated by the challenge of driving a mobile robot on an extremely slippery surface with high-accuracy motion for curling. To prevent slippage, a slip ratio controller is adopted based on the difference between the wheel and robot speeds. To maintain a desired heading angle of the mobile robot during driving, orientation control based on the yaw dynamics is developed. Moreover, the robot does not require any positioning information from external sources, such as sensors and cameras. In addition to playing an actual game, this curling robot can be utilized for training athletes. Furthermore, the developed algorithms for high-accuracy driving on ice in an indoor stadium can be applied to various mobile robots that drive on other slippery environments (such as sand and dust) and sealed spaces (such as underground warehouses) where external positioning signals cannot be received. The proposed robot is limited to straight driving, and the possibility of extending it to circular motion is still under development. IEEE