Method for acquiring ball-hitting gesture and ball-hitting speed of ping-pong robot racket

Abstract

The invention discloses a method for acquiring a ball-hitting gesture and a ball-hitting speed of a ping-pong robot racket. The method comprises the following steps of: establishing a ping-pong ball flying analysis model; calculating a ball outlet speed after the robot racket hits a ball by using the model according to positions of an expected ball-returning falling point and a predicted hitting point in combination with a speed of the expected ball-returning falling point; deriving a group of non-linear equations containing the ball-hitting gesture and the ball-hitting speed of the racket according to the speeds of a ping-pong ball before and after the racket hits the ball and in combination with a rebounding model of the racket and a kinematics model of a ping-pong robot; and solving the non-linear equations by using a known optimization method, and calculating the ball-hitting gesture and the ball-hitting speed of the ping-pong robot racket.

Description

technical field [0001] The invention belongs to motion planning in the field of robot control, in particular to a method for obtaining the batting posture and batting speed of a table tennis robot racket. Background technique [0002] Traditionally, the calculation of batting attitude and batting speed of table tennis robot rackets is mainly realized by the following two methods: 1) Simplify the table tennis flight trajectory model into a parabola, and simplify the racket bounce model into a specular reflection model (see "Kalpesh P.M., Ferat S., and Eli S., An application of human robot interaction: development of a ping-pong playing robotic arm IEEE International Conference on Systems, Man and Cybernetics, Vol. 2, pp. 1831-1836, 2005.”). 2) Utilize learning algorithm and expert system, obtain racket batting posture and batting speed (see " L.Acosta, J.Rodrigo, J.A.Mendez, G.N.Marichal and M.Sigut, Ping-pong player prototype, a PC-based, low-Cost, ping pong robot, IEEE Rob...

AI Technical Summary

Problems solved by technology

However, there are large deviations in the batting posture and batting speed calculated by the model, resulting in scattered landing points when the table tennis robot returns the ball, and poor quality of the ball returned. beat

The second method uses a learning algorithm and an expert system to obtain the batting posture and speed of the racket through a large amount of training data. For the data in the sampled data set, the table tennis robot can calculate the correct batting posture and speed of the racket, but for For data other than the sampling data set, the racket hitting posture and hitting speed calculated by the robot cannot meet the control accuracy requirements

Moreover, when training the sampling data, various factors such as the speed of the ball and the position of the ball jointly determine the position and speed of the table tennis robot's return ball. The training is more difficult, and the position and speed of the return are uncontrollable, which affects the athlete. training quality

Images