Pass control method of humanoid soccer robots and team ball-control method

A football robot and control method technology, applied in the field of football robots, can solve the problems of little significance and failure to take into account the cooperation of players, etc., and achieve the effect of increasing the holding time of the ball, improving the success rate of passing the ball, and increasing the winning rate

CN106964145AActive Publication Date: 2017-07-21NANJING UNIV OF POSTS & TELECOMM
6 Cites 20 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2017-07-21

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to a pass control method of humanoid soccer robots and a team ball-control method. By the adoption of brand new training model design, control is conducted regarding a pass motion which is determined after a robot holding a ball determines to pass the ball, and the success rate of pass can be effectively increased. On the basis of the pass control method of the humanoid soccer robots, the team ball-control method is further designed, overall-planning and control are conducted regarding the robots in a whole team; the success rate of the pass is increased, the ball-holding duration of the team can be effectively prolonged at the same time, and the winning rate is increased.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to a humanoid soccer robot pass control method and a team ball control method, belonging to the technical field of soccer robots. Background technique

[0002] Robot soccer is a frontier topic in the field of artificial intelligence and robot research. The RoboCup 3D simulation environment is based on a general multi-agent physical simulator - simspark. The simulator uses the open dynamics engine (OpenDynamics Engine ODE) library to achieve its realistic simulation of rigid body dynamics and collision detection and friction. With the development of robot technology, effective passing action and passing tactics have become a hot issue in this field.

[0003] In order to solve the robot's passing problem in the 2D simulation platform, Peter stone proposed an effective training model - KeepAway. This training model is a typical reinforcement learning model. Its training idea is to train a fixed number of players on both sides in a...

Examples

Embodiment Construction

[0047]The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0048] In the design process of the humanoid soccer robot, in order to use the KeepAway training model on the RoboCup3D simulation platform to train effective passing tactics, it is necessary to improve the training model and learning algorithm. According to the KeepAway training idea and with reference to the training model in the RoboCup2D simulation platform, the present invention designs a humanoid soccer robot passing control method, which controls the passing action after the ball-holding robot determines the passing. In practical applications, Such as figure 1 As shown, it specifically includes the following steps:

[0049] Step 001. Determine whether the number of own robots within the field of view of the robot holding the ball is greater than 1, if yes, go to step 003; otherwise, go to step 002.

[0050] Step ...