Fuzzy Controller based Biped Robot Balance Control using 3D Image

In this paper, we propose a balance control algorithm based on fuzzy controller using 3D (Dimensional) geometric information obtained from sequential images. The pose of a biped robot is acquired from image information instead of different sensors, and this pose information is used for fuzzy controller to keep biped robot in a stable balanced condition. Initially, the camera mounted on the head of robot obtains the pose information of the target object while walking. We acquired corresponding points between characteristic features on the target object and CAD (Computer Aided Design) information in DB(Data Base) using SURF (Speeded Up Robust Features) algorithm. Three arbitrary features from SURF algorithm are able to generate a virtual plane, which represent the geometric information of the target object’s pose data. The camera pose information for the target object is calculated using pose information derived from the virtual planes. As the camera pose information is related to pose of a biped robot, we can easily get the pose information of a biped robot. With this the difference between current and desired pose of a biped robot is obtained. This difference is used as input data for the fuzzy controller. The fuzzy controller keeps the biped robot in stable pose using the recent pose and velocity of a biped robot. The efficiency of the proposed algorithm has been proven by the experiments performed on even floor using a mini humanoid robot. technological development facilitated the advancement in the areas which has resulted in a number of successful systems. Such applications are mostly concentrated on visual servoing of manipulators which controls the pose of end-effecter to approach a specific trajectory by measuring the target trajectory using the visual information [1, 2]. Since visual servoing is used to control a manipulator based on analysis of images being provided in real-time, Chesi [3] proposed a method of using 6 instead of 8 feature points which have been required for the traditional visual servoing techniques to extract the 3D pose of the reference object. Radu [4] proposed a method of aligning the pose of the first manipulator (where a camera is mounted on its end-effecter) to the pose of the second manipulator using the visual information provided by the camera. The pose of the first manipulator is estimated from the image feature points of the second manipulator’s gripper. There are also few other examples that have applied visual servoing techniques for balance control of biped robots. Different from the visual servoing techniques used for the manipulators whose base frame is fixed to the ground, but those which are used in biped robots, should consider the fact that the COM(Center Of Mass) or the ZMP (Zero Moment Point) of robot is not static while robot is walking. In order to apply the visual servoing technique used for manipulator having 6 DOF (Degrees Of Freedom) to the positioning control of a biped robot, Yamamura [5] fixed the geometrical relation between the camera and the robot frames to reduce the degree of freedom of each leg to 4. This approach of adapting a visual servoing technique for manipulator to the biped robot limits the robot’s locomotion significantly. With stereo vision, we can see where the objects are in relation to our own bodies with much greater precision, especially when those objects are moving toward or away from us in the depth dimension. Stereovision has been studied long, and common computational approaches include feature-based [6], area-based methods [7]. All these methods have their intrinsic problems, caused by the very assumptions inherent in these approaches. These methods, by using multiple cameras such as stereo vision are able to get more accurate and stable 3D information than single camera method. But, it is not a more economic way of finding the depth of all objects or a view.