计算机工程与应用 ›› 2021, Vol. 57 ›› Issue (7): 136-143.DOI: 10.3778/j.issn.1002-8331.1912-0337

• 模式识别与人工智能 • 上一篇    下一篇

结合Winner-Take-All的足球机器人控制策略研究

廖列法,李浩瀚,李帅,朱合隆,李志军   

  1. 1.江西理工大学 信息工程学院,江西 赣州 341000
    2.斯旺西大学 工程学院,英国 威尔士 斯旺西 SA1 8EN
  • 出版日期:2021-04-01 发布日期:2021-04-02

Research on Control Strategy of Soccer Robot Combined with Winner-Take-All

LIAO Liefa, LI Haohan, LI Shuai, ZHU Helong, LI Zhijun   

  1. 1. College of Applied Science, Jiangxi University of Science and Technology, Ganzhou, Jiangxi 341000, China
    2. College of Engineering, Swansea University, Singleton Park, Swansea, Wales,SA1 8EN, UK
  • Online:2021-04-01 Published:2021-04-02

摘要:

在足球机器人运动过程中,足球机器人处于一个实时对抗的复杂环境中,这就需要机器人有较高的实时运动过程应对能力。需要对每个关键时刻,例如:多机器人抢球过程、单机器人控球过程等,做出合理的应对措施。许多策略的研究都只注重单机器人控球过程的路径规划,没有考虑到多机器人竞争的过程,导致足球机器人整个运动过程中的一些关键步骤的缺失,丧失了完整性,忽略了实时的对抗性。拟采用新的策略解决上述问题:第一步是将采用WTA(Winner Take All)竞争模型去有效的解决多机器人竞争问题;第二步将采用一种改进的APF(Artificial Potential Field)路径规划法来进行避障。解决了传统APF算法的弊端,提高了效率。通过仿真实验,验证了理论的正确性,也验证了所提理论的科学性和实用性,为以后在其他科学领域的实践奠定了基础。

关键词: 足球机器人, WTA竞争模型, 人工势场(APF), 路径规划, 避障

Abstract:

In the movement process of soccer robot, the soccer robot is in a complex real-time confrontation environment, which requires the robot to have a high ability to deal with the real-time movement process and to make reasonable countermeasures for each critical moment, such as multi-robot ball grabbing process, single-robot ball control process, etc.. Many strategy studies only focus on the path planning of the ball control process of a single robot but fail to consider the process of multi-robot competition, resulting in the absence of some key steps in the whole movement process of soccer robot, loss of integrity, and neglect of real-time antagonism. A new strategy is proposed to solve the appeal problem:The first step is to use the WTA(Winner Take All) competition model to effectively solve the multi-robot competition problem; the second will adopt an improved APF(Artificial Potential Field) path planning method to avoid obstacles. Thus, the disadvantages of traditional APF algorithm are solved and the efficiency is improved. Through the simulation experiment, this paper verifies the correctness, scientific rigor and practicability of the theory. It lays a foundation for the practice in other scientific fields.

Key words: soccer robot, Winner Take All(WTA) competition model, Artificial Potential Field(APF), path planning, obstacle avoidance