Motion data generating system for robot, motion data generating method for robot, and storage medium

A motion data generating method for a robot provided with at least a waist and a leg, the method comprising:

• acquiring captured leg motion data captured from a motion of the separating leg of a test subject and captured waist motion data captured from a motion of the waist;
• smoothing the captured leg motion data, performing peak detection, and extracting at least a separation starting point, a peak position, and a separation ending point as representative points;
• converting the captured leg motion data and the captured waist motion data at the representative point to a scale of the robot;
• setting up a target zero moment point (ZMP) of the robot at the representative point;
• generating postures of the leg and the waist of the robot for every step in accordance with a control cycle of the robot between the representative points by interpolation;
• generating motion data in which the postures of the leg and the waist of the robot for each step are modified so as to satisfy the target ZMP;
• performing the peak detection individually with respect to a position of three-dimensional coordinates of a foot tip in a x-direction, in a y-direction, and in a z-direction, and also for an angle of a roll axis, an angle of a pitch axis, and an angle of a pitch axis;
• and
• adding all detected representative points, and performing thinning on the representative points that have a plurality of time widths at intervals equal to or less than a predetermined time interval.

A non-transitory storage medium storing a motion data generating program for a robot provided with at least a waist and a leg, the motion data generating program executed by a computer operating as an arithmetic device, comprising:

• acquiring captured leg motion data captured from a motion of the separating leg of a test subject and captured waist motion data captured from a motion of the waist;
• smoothing the captured leg motion data, performing peak detection, and extracting at least a separation starting point, a peak position, and a separation ending point as representative points;
• converting the captured leg motion data and the captured waist motion data at the representative point to a scale of the robot;
• setting up a target zero moment point (ZMP) of the robot at the representative point;
• generating postures of the leg and the waist of the robot for every step in accordance with a control cycle of the robot between the representative points by interpolation;
• generating motion data in which the postures of the leg and the waist of the robot for each step are modified so as to satisfy the target ZMP;
• performing the peak detection individually with respect to a position of three-dimensional coordinates of a foot tip in a x-direction, in a y-direction, and in a z-direction, and also for an angle of a roll axis, an angle of a pitch axis, and an angle of a pitch axis;
• and
• adding all detected representative points, and performing thinning on the representative points that have a plurality of time widths at intervals equal to or less than a predetermined time interval.