# US 11,305,161 B2 · Intelligent ball passing system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,305,161 B2 - Document ID: US11305161B2 - Status: Granted - Kind: B2 - Application number: 16/753,008 - Filed: 2018-06-01 - Published: 2020-07-23 - Granted: 2022-04-19 - Earliest claimed: 2017-11-21 - Assignees: Wuyi University - Applicants: Wuyi University - Inventors: Tianlei Wang; Weitian Ni; Liping Xu; Rong Fu; Hongwei Yue; Bing Luo; Zhanpeng Zhao; Shuqing Lei - Web page: https://patentgrep.com/patent/US11305161B2 ## CPC - A63B 47/002: {Devices for dispensing balls, e.g. from a reservoir (automatic teeing devices A63B57/0006; devices for projecting balls A63B69/40)} - G06N 3/084: Backpropagation, e.g. using gradient descent - A63B 24/0062: {Monitoring athletic performances, e.g. for determining the work of a user on an exercise apparatus, the completed jogging or cycling distance} - G06V 40/23: {Recognition of whole body movements, e.g. for sport training} - A63B 47/021: {for picking-up automatically, e.g. by apparatus moving over the playing surface} - A63B 69/38: for tennis {(A63B61/006, A63B69/0073 and A63B69/0097 take precedence)} - A63B 2047/022: {Autonomous machines which find their own way around the playing surface} - A63B 2102/02: Tennis - A63B 2220/05: Image processing for measuring physical parameters - A63B 2220/80: Special sensors, transducers or devices therefor - A63B 2220/806: Video cameras - A63B 2220/833: Sensors arranged on the exercise apparatus or sports implement - A63B 2225/50: Wireless data transmission, e.g. by radio transmitters or telemetry - A63B 2230/62: posture - A63B: APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT (apparatus for passive exercising, massage A61H) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract An intelligent ball passing system is provided, including: a ball passing robot configured for carrying out a ball passing action according to human body postures, including: a video capture device configured for acquiring human body posture data, a movement mechanism configured for driving the ball passing robot to move according to a ball passing instruction, and a central processing module connected to the movement mechanism and the video capture device respectively; and a software system configured for controlling the ball passing robot to carry out the ball passing action, the central processing module recognizing and classifying the human body postures by using the software system that uses a BP neural network. ## Claims ### Claim 1 An intelligent ball passing system, comprising: a ball passing robot configured for carrying out a ball passing action according to human body postures, comprising: a video capture device configured for acquiring human body posture data, a movement mechanism configured for driving the ball passing robot to move according to a ball passing instruction, and a central processing module connected to the movement mechanism and the video capture device respectively, and a software system configured for controlling the ball passing robot to carry out the ball passing action, the central processing module recognizing and classifying the human body postures by using the software system that uses a BP neural network. ### Claim 2 (depends on 1) The intelligent ball passing system of claim 1, wherein recognizing and classifying the human body postures by using the software system that uses a BP neural network comprises: A. acquiring, by the video capture device, a posture picture corresponding to the human body posture data; B. performing human body separation on the posture picture by using an OPENCV library to obtain a separated picture; C. determining whether the separated picture corresponds to a human body, binarizing the separated picture and outputting a binarized image if the separated picture corresponds to a human body, and stopping recognition and classification operation if the separated picture does not correspond to a human body; D. training the BP neural network, inputting the binarized image into the trained BP neural network, and extracting a human body posture feature value; and E. recognizing and classifying the human body posture feature value, driving, by the movement mechanism, the ball passing robot to move and pass a ball if the human body posture feature value corresponds to a ball passing instruction, and having the ball passing robot standby if the human body posture feature value does not correspond to a ball passing instruction. ### Claim 3 (depends on 2) The intelligent ball passing system of claim 2, wherein, in step B, during human body separation from the posture picture by using the OPENCV library, the human body is separated out from the posture picture by Gaussian background modeling. ### Claim 4 (depends on 1) The intelligent ball passing system of claim 1, wherein the movement mechanism comprises wheels and a motor for driving the wheels, and the wheels are arranged on the bottom of the ball passing robot. ### Claim 5 (depends on 1) The intelligent ball passing system of claim 1, wherein the ball passing robot further comprises an obstacle avoidance sensor configured for providing the ball passing robot with obstacle conditions in front thereof and connected to the central processing module. ### Claim 6 (depends on 1) The intelligent ball passing system of claim 1, wherein the ball passing robot further comprises a wireless communication module configured for the ball passing robot to communicate with an external terminal and connected to the central processing module. ### Claim 7 (depends on 1) The intelligent ball passing system of claim 1, wherein the ball passing robot further comprises a lithium battery pack configured for providing the ball passing robot with an operating power source. ### Claim 8 (depends on 1) The intelligent ball passing system of claim 1, wherein the central processing module is composed of a processing chip S5P6816 and a peripheral circuit thereof. ### Claim 9 (depends on 1) The intelligent ball passing system of claim 1, wherein the video capture device is a camera.