Ball Tracking and Trajectory Prediction for Table-Tennis Robots

Abstract

Sports robots have become a popular research topic in recent years. For table-tennis robots, ball tracking and trajectory prediction are the most important technologies. Several methods were developed in previous research efforts, and they can be divided into two categories: physical models and machine learning. The former use algorithms that consider gravity, air resistance, the Magnus effect, and elastic collision. However, estimating these external forces require high sampling frequencies that can only be achieved with high-efficiency imaging equipment. This study thus employed machine learning to learn the flight trajectories of ping-pong balls, which consist of two parabolic trajectories: one beginning at the serving point and ending at the landing point on the table, and the other beginning at the landing point and ending at the striking point of the robot. We established two artificial neural networks to learn these two trajectories. We conducted a simulation experiment using 200 real-world trajectories as training data. The mean errors of the proposed dual-network method and a single-network model were 39.6 mm and 42.9 mm, respectively. The results indicate that the prediction performance of the proposed dual-network method is better than that of the single-network approach. We also used the physical model to generate 330 trajectories for training and the simulation test results show that the trained model achieved a success rate of 97% out of 30 attempts, which was higher than the success rate of 70% obtained by the physical model. A physical experiment presented a mean error and standard deviation of 36.6 mm and 18.8 mm, respectively. The results also show that even without the time stamps, the proposed method maintains its prediction performance with the additional advantages of 15% fewer parameters in the overall network and 54% shorter training time.

Learning from machine learning: prediction of age-related athletic performance decline trajectories

“…Particularly, the prediction of the performance decline trajectory of an individual from only one measurement would be desirable, as it permits an immediate assessment without requiring results from several previous years that can often not be obtained. ML approaches were successful in the prediction of septic shock onset [ 18 ], epileptic seizures [ 19 ], the onset of type 2 diabetes mellitus [ 20 ], or ball trajectories for table-tennis robots [ 21 ]. In sports, the prediction of the potential and performance trajectories of young talents to identify future champions by ML has been demonstrated for archers [ 22 ] and in table tennis [ 23 ].…”

A Comparative Analysis Using Silhouette Extraction Methods for Dynamic Objects in Monocular Vision

“…Several studies already exist on rotational detection. In [24], a method for detecting the rotation using a ball marked with some feature points was provided and the rebound phenomenon between a ball and the table/racket rubber was also studied therein. High-speed multiple cameras with the capturing rate of 900 frames per second (f/s) or 1200 f/s were also needed to capture the image of the ball with marks [25][26].…”

Ball Motion Control in the Table Tennis Robot System Using Time-Series Deep Reinforcement Learning

“…For example, the table tennis robot studied in this research has been designed like a humanoid system, which can sense the position, speed and spin velocity of the incoming ball, and then make the decision of stroke strategy followed by the controlling the robot to stroke back the ball. In such a case, building an accurate physical model of the table tennis environment is essential for designing the robot control system [2], [3]. Table tennis can be considered as a complicated physical system including the ball flying physical model, collision model, racket hit model.…”

“…Based on our testing result, during the table tennis game, the ball flying speed and spin velocity can reach as high as 25 m/s and 300 rad/s respectively. Although many studies have been made on the ball tracking and trajectory prediction of the table tennis [3]- [5], seldom attempts had been made for the ball's spinning velocity estimation. Although obtaining an accurate spin velocity is a crucial step for the stroke strategy decision of the table tennis robot to play and compete with human players, it is very difficult to capture the spin velocity of the ball using a regular vision based method.…”