System and method for motion control of humanoid robot

A humanoid robot, comprising:

• a plurality of drive motors for varying angles of a plurality of joints of the robot;
• a three-dimensional camera for recognizing three-dimensional image information, which includes two-dimensional image information and distance information of a user;
• an image processor for calculating a variation in angles of the joints based on three-dimensional coordinates of a first reference point and a second reference point based on the three-dimensional image information to generate joint control data;
• and
• a controller for checking the joint control data and transmitting motor control signals to respective drive motors by comparing the joint control data with previous angles of joints and dividing the joint control data into time-based data through interpolation.

A method for remotely controlling a robot, comprising:

• recognizing, by a remote control device, three-dimensional information, which includes two-dimensional image information and distance information of a user;
• determining three-dimensional coordinates of a first reference point from the three-dimensional image information;
• determining three-dimensional coordinates of a second reference point from the first reference point;
• calculating a variation in angles of joints of the robot based on the three-dimensional coordinates of the first reference point and the second reference point;
• transmitting a joint control signal, which includes joint control data according to the variation in angles of the joints, to the robot through a wired/wireless network;
• and
• checking, by the robot, joint control data from the joint control signal received from the remote controller, and transmitting motor control signals to respective drive motors by comparing the joint control data with previous angles of joints and dividing the joint control data into time-based data through interpolation.

The method according to 13, wherein determining the three-dimensional coordinates of the first reference point comprises:

• removing a background image from the three-dimensional image information;
• extracting a silhouette of the user's image and determining a head center point of the user in accordance with feature points;
• and
• calculating an inclination of a feature line formed by connecting the feature points to determine the first reference point.

A method for motion control of a robot, comprising:

• recognizing, by the robot, three-dimensional image information, which includes two-dimensional image information and distance information of a user;
• determining three-dimensional coordinates of a first reference point in accordance with the three-dimensional image information;
• determining three-dimensional coordinates of a second reference point based on the first reference point;
• calculating a variation in angles of joints of the robot based on the three-dimensional coordinates of the first reference point and the second reference point;
• and
• checking joint control data from a joint control signal and transmitting motor control signals to respective drive motors by comparing the joint control data with previous angles of joints and dividing the joint control data into time-based data through interpolation.