基于神经网络的撞球机器人控制器设计

doi:10.13700/j.bh.1001-5965.2016.0183

北京航空航天大学学报 ›› 2017, Vol. 43 ›› Issue (3): 533-543.doi: 10.13700/j.bh.1001-5965.2016.0183

基于神经网络的撞球机器人控制器设计

高家颖¹, 何秋阳², 詹志新¹

1. 北京航空航天大学航空科学与工程学院, 北京 100083；;
2. 约克大学计算机科学学院, 约克 YO10 5DD

收稿日期:2016-03-09 出版日期:2017-03-20 发布日期:2016-10-10
通讯作者: 何秋阳,E-mail:heqiuyang926@163.com E-mail:heqiuyang926@163.com
作者简介:高家颖,男,博士研究生。主要研究方向:机器学习。;何秋阳,女,信息安全工程师。主要研究方向:信息安全、人机交互。

Design of neural network controller for a billiard robot

GAO Jiaying¹, HE Qiuyang², ZHAN Zhixin¹

1. School of Aeronautic Science and Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100083, China;
2. Department of Computer Science, The University of York, York YO10 5DD, UK

Received:2016-03-09 Online:2017-03-20 Published:2016-10-10

摘要/Abstract

摘要： 对撞球机器人的母球控制问题展开研究，设计了一种基于神经网络（NN）的控制器，使机器人能够控制母球在击打目标球后按照预定的模式运动至目标点——即完成走位。针对该问题非线性且非光滑的特点，对坐标系进行阐述并给出机器人击球的模型；在光滑的假设下使用理论分析的方法建立母球的运动学模型与边库反弹的理想镜像模型；进而使用神经网络方法对理想模型进行修正，并对不同的轨迹模式进行分析与分类。测试结果表明：经过训练的机器人能够掌握各种模式的走位，统计结果与模型分析结果相吻合；相比于单一使用神经网络方法，本文使用理论分析与神经网络相结合的方法能够有效地提升网络的品质，降低训练的误差。

关键词: 撞球机器人, 运动学分析, 神经网络(NN), 轨迹分类器, 非线性非光滑

Abstract: This paper focuses on the cue ball controlling problem for a billiard robot. A neural network (NN) controller is designed, and the trained robot is able to stroke the cue ball moving to the target point after colliding with objective ball and cushions. Since the problem is non-linear and non-smooth, the solution is divided into several steps. First, the stroking model and the coordinate definition are described. Second, the kinematic model for cue ball motion and the mirror model for cushion rebounds are established under the ideal smooth assumption. Then, the neural network method is used to modify the ideal models, and the pattern recognition method for trajectories is presented. In the verification test, the trained robot is able to master the cue ball controlling with each pattern. The statistic results tally with the model analysis. Compared with simply adopting neural network method, the method combined with theoretical kinematic analysis will effectively improve the network quality and reduce the training error.

Key words: billiard robot, kinematic analysis, neural network (NN), trajectory classification, non-linear and non-smooth

中图分类号:

TP183

高家颖, 何秋阳, 詹志新. 基于神经网络的撞球机器人控制器设计[J]. 北京航空航天大学学报, 2017, 43(3): 533-543.

GAO Jiaying, HE Qiuyang, ZHAN Zhixin. Design of neural network controller for a billiard robot[J]. JOURNAL OF BEIJING UNIVERSITY OF AERONAUTICS AND A, 2017, 43(3): 533-543.

参考文献

[1] HO K H L,MARTIN T,BALDWIN J.Snooker robot player-20 years on[C]//IEEE Symposium on Computational Intelligence and Games,2007.(CIG 2007).Piscataway,NJ:IEEE Press,2007:1-8.
[2] YANG J S,CHEN Y M.The design of jump shot decision-making system for a billiard robot[C]//2010 IEEE Conference on Robotics Automation and Mechatronics(RAM).Piscataway,NJ:IEEE Press,2010:409-413.
[3] CHENG B R,LI J T,YANG J S.Design of the neural-fuzzy compensator for a billiard robot[C]//2004 IEEE International Conference on Networking,Sensing and Control.Piscataway,NJ:IEEE Press,2004:909-913.
[4] LEGG P A,PARRY M L,CHUNG D H S,et al.Intelligent filtering by semantic importance for single-view 3D reconstruction from Snooker video[C]//IEEE International Conference on Image Processing.Piscataway,NJ:IEEE Press,2011:2385-2388.
[5] LECKIE W,GREENSPAN M A.Pool physics simulation by event prediction 1:Motion transitions[J].ICGA Journal,2005,28(4):214-222.
[6] LANDRY J F,DUSSAULT J P,MAHEY P.A heuristic-based planner and improved controller for a two-layered approach for the game of billiards[J].IEEE Transactions on Computational Intelligence & AI in Games,2013,5(4):325-336.
[7] DOMÉNECH A.Non-smooth modelling of billiard- and superbilliard-ball collisions[J].International Journal of Mechanical Sciences,2008,50(4):752-763.
[8] MATHAVAN S,JACKSON M R,PARKIN R M.Numerical simulations of the frictional collisions of solid balls on a rough surface[J].Sports Engineering,2014,17(4):227-237.
[9] MATHAVAN S,PARKIN R M,JACKSON M R.A theoretical analysis of billiard ball dynamics under cushion impacts[J].Proceedings of the Institution of Mechanical Engineers,Part C:Journal of Mechanical Engineering Science,2010,224(9):1863-1873.
[10] GAO J,HE Q,ZHAN Z,et al.Dynamic modeling based on fuzzy neural network for a billiard robot[C]//IEEE International Conference on Networking,Sensing and Control.Piscataway,NJ:IEEE Press,2016:1-4.
[11] ZUO Z,WANG C.Adaptive trajectory tracking control of output constrained multi-rotors systems[J].IET Control Theory & Applications,2014,8(13):1163-1174.
[12] LOU W,GUO X.Adaptive trajectory tracking control using reinforcement learning for quadrotor[J/OL].International Journal of Advanced Robotic Systems,2016,13(38).[2016-03-09].http://cdn.intechopen.com/pdfs-wm/49981.pdf.DOI:105 772/62128.
[13] LECKIE W,GREENSPAN M A.Pool physics simulation by event prediction 2:Collisions[J].ICGA Journal,2006,29(1):24-31.

基于神经网络的撞球机器人控制器设计

Design of neural network controller for a billiard robot

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 10

编辑推荐

Metrics

本文评价

[1]	房海蓉, 王立, 张海强, 杨会. 一种含闭环支链的新型并联机构设计与分析[J]. 北京航空航天大学学报, 2019, 45(3): 454-463.
[2]	杨会, 房海蓉, 李典, 方跃法. 一种新型并联灌注机器人运动学分析和多目标优化[J]. 北京航空航天大学学报, 2018, 44(3): 568-575.
[3]	李典, 郭盛, 黄冠宇, 赵福群. 新型2-2PRUR并联机构运动学分析及工作空间优化[J]. 北京航空航天大学学报, 2017, 43(10): 2011-2020.
[4]	丁建中, 王春洁. 含铰链间隙板式卫星天线展开精度分析[J]. 北京航空航天大学学报, 2016, 42(12): 2625-2631.
[5]	史震云, 朱前成. 基于智能驱动器的软体机器人系统[J]. 北京航空航天大学学报, 2016, 42(12): 2596-2602.
[6]	常乐, 刘正华, 温暖, 吴森堂. 新型三轴离心机系统构型及数学建模[J]. 北京航空航天大学学报, 2015, 41(2): 283-288.
[7]	张利格, 毕树生, 彭朝琴. 空间四面体翻滚机器人运动学分析及仿真实验[J]. 北京航空航天大学学报, 2011, 37(4): 415-420.
[8]	丁希仑, 周乐来, 周军. 机器人的空间位姿误差分析方法[J]. 北京航空航天大学学报, 2009, 35(2): 241-245.
[9]	贾明, 毕树生, 宗光华, 徐一村. 仿生扑翼机构的设计与运动学分析[J]. 北京航空航天大学学报, 2006, 32(09): 1087-1090.
[10]	任敬轶, 孙汉旭. 自由漂浮空间机器人的运动学分析及仿真[J]. 北京航空航天大学学报, 2002, 28(6): 613-616.