作者：王琳玮, 邵星灵, 杨卫

作者单位：中北大学 电子测试技术国家重点实验室，山西 太原 030051

关键词：球形机器人;重心偏离;神经网络;PID算法;惯性传感器

摘要：

针对复杂环境下对机器人多自由度灵活稳定运动的需求，设计一种基于神经网络自整定PID控制算法的球形机器人；该系统采用重心偏离的方式为机器人提供动力，利用对质量块的智能调速改变机器人的方向，结合陀螺仪加速度计等惯性传感器实现对机器人方向和位移的精确控制；机器人内部安装包括图像采集系统在内的多种传感器，利用设计的遥控程序，可以通过电脑或手机APP功能通过WiFi实现机器人远程控制。对机器人性能进行测试，实验结果表明该机器人运动性能良好，运动轨迹和姿态角误差在控制器接受范围内，在管道、密闭空间探查等方面具有良好应用前景。

Experimental research of spherical robot pose control system based on inertial sensor 
WANG Linwei, SHAO Xingling, YANG Wei
State Key Laboratory of Electronic Testing Technology, North University of China, Taiyuan 030051, China
Abstract: Aiming at complex environment more degrees of freedom robot flexible stable movement of the demand, design a self-tuning PID control algorithm based on neural network of the spherical robot. The system use a center of gravity deviation to provide power for robot, using the change of mass of intelligent control for the direction of the robot, in combination with gyroscope inertial sensors such as accelerometer to realize accurate control of the direction and displacement of the robot. Robot installed inside, including image acquisition system, a variety of sensors, Remote control procedures, can be remotely controlled by WiFi through the function of computer or mobile APP. The performance of the robot was tested, and the experimental results showed that the robot had good motion performance, the motion trajectory and attitude angle errors are within the range accepted by the controller and have a good application prospect in pipeline confined space exploration
Keywords: spherical robot;center of gravity deviation;neural net;PID algorithm;inertial sensor
2020, 46(3):123-127  收稿日期: 2019-07-09;收到修改稿日期: 2019-08-27
基金项目: 国家自然科学基金（61803347）；山西省应用基础研究项目（201701D221123）
作者简介: 王琳玮(1993-),男,黑龙江绥化市人,硕士研究生,专业方向为异构机器人协同控制
参考文献
[1] 战强, 李伟. 球形移动机器人的研究进展与发展趋势[J]. 机械工程学报, 2019, 55(9): 1-17
[2] 陆震. 人工智能在军用机器人的应用[J]. 兵器装备工程学报, 2019, 40(5): 1-5
[3] 于涛, 杨昆, 赵伟. 一种球形机器人的自适应鲁棒轨迹跟踪控制[J]. 中国测试, 2018, 44(3): 102-108
[4] 刘增波, 战强, 蔡尧. 一种环境探测球形移动机器人的运动控制[J]. 航空学报, 2008, 29(6): 1673-1679
[5] ZHAO W, KAMEZAKI M, YOSHIDA K, et al. A coordinated wheeled gas pipeline robot chain system based on visible light relay communication and illuminance assessment.[J]. Sensors (Basel, Switzerland), 2019, 19(10): 2322
[6] 兰晓娟, 孙汉旭, 贾庆轩. 水下球形机器人BYSQ-2的原理与动力学分析[J]. 北京邮电大学学报, 2010, 33(3): 20-23,29
[7] 杨伟, 李健, 肖起阳, 等. 球形机器人的仿真与实验测试分析[J]. 制造业自动化, 2017, 39(8): 26-29,34
[8] PAN D, GAO F, MIAO Y, et al. Co-simulation research of a novel exoskeleton-human robot system on humanoid gaits with fuzzy-PID / PID algorithms[J]. Advances in Engineering Software, 2015, 79(C): 36-46
[9] WANG Z, GUO S, SHI L, et al. The application of PID control in motion control of the spherical amphibious robot[C]//2014 IEEE International Conference on Mechatronics and Automation. IEEE, 2014: 1901-1906.
[10] HERNANDEZ A R, GARCIA L G, SALGADO J T. Neural network-based self-tuning PID control for underwater vehicles.[J]. Sensors (Basel, Switzerland), 2016, 16(9): 1429
[11] 吴晗, 薛磊, 徐开芸, 等. 球形机器人双闭环PID控制设计与仿真[J]. 机电产品开发与创新, 2018, 7(4): 82-84
[12] EOROV I, ZIMIN M A. Nonlinear correction of bilateral remote control systems within a mobile robot pipeline[J]. Procedia Computer Science, 2017, 103: 522-527