传统控制方法过分依赖于机器人运动力学，在流动海平面上容易出现偏航，为了解决这个问题，提出了舰船导航机器人的自动控制研究。使用激光雷达辨识障碍物边缘和轮廓信息，通过构建直角坐标系，确定目的地的位置信息。设计最优反馈控制器结构，构建最优控制函数，结合线性二次最优控制理论，使反馈最优控制函数值取得最小值，获取舰船导航机器人的自动控制规律，使系统控制具有动态响应性能。采用动态窗口法控制导航机器人避障功能，构建运动学模型，将输入偏差信号转换为论域上的点，通过设定合理阈值实现控制器平滑切换，推导出运动轨迹方程，实现对导航方向的自动控制。实验结果表明，该方法能够按照既定路线精准避障，角度偏差控制在−5°～5°范围内，位移偏差x，y，z方向分别控制在0～0.25 m，−1.0～0 m，−0.25～0 m范围内，能够达到自动精准控制的目的。
Traditional control methods rely too much on robot motion mechanics, which is prone to yaw on flowing sea level. In order to solve this problem, the automatic control of ship navigation robot is proposed. The edge and contour information of the obstacle are obtained by liDAR identification, and the location information of the destination is determined by constructing a rectangular coordinate system. The structure of the optimal feedback controller is designed, and the optimal control function is constructed. Combined with the linear quadratic optimal control theory, the value of the feedback optimal control function is minimized, and the automatic control law of the ship navigation robot is obtained, so that the system control has dynamic response performance. The dynamic window method was used to control the obstacle avoidance function of the navigation robot, and the kinematics model was constructed. The input deviation signals were converted into points on the domain. The controller was smoothly switched by setting a reasonable threshold, and the motion trajectory equation was derived to realize the automatic control of the navigation direction. The experimental results show that the method can accurately avoid obstacles according to the established route, the Angle deviation is controlled within −5°～5°, and the displacement deviation x, y and z directions are controlled within 0～0.25 m, −1.0～0 m and −0.25～0 m, respectively, which can achieve the purpose of automatic and accurate control.
2022,44(19): 138-141 收稿日期：2022-07-04
DOI：10.3404/j.issn.1672-7649.2022.19.027
分类号：TP273
作者简介：吕世霞(1978-)，女，硕士，副教授，研究方向为机器人技术应用及自动控制技术
参考文献：
[1] 丁坤, 王立, 张经炜, 等. 基于改进导航圆算法的光伏阵列清扫机器人路径跟踪控制[J]. 河海大学学报(自然科学版), 2021, 49(2): 190–196
[2] 李宁. RTK与Lidar组合优化机器人导航控制[J]. 机械设计与制造, 2021(11): 275–278
[3] 包加桐, 姚小梅, 缪露, 等. 基于云平台地图的侦察机器人室外自主导航[J]. 控制工程, 2020, 27(6): 941–946
[4] 王钰奉, 陈伟堤, 施祥玲, 等. Delta机器人增广非线性PD控制研究[J]. 组合机床与自动化加工技术, 2021(12): 62–66
[5] 刘冬, 陈飞, 邹强, 等. 基于情景经验与稀疏点云的移动机器人导航[J]. 华中科技大学学报(自然科学版), 2020, 48(9): 25–30
[6] 张玉超, 蒋沅, 代冀阳. 三阶多机器人协同编队动态避障控制[J]. 系统仿真学报, 2022, 34(8): 1762–1774
[7] 李航, 王朝耀. 仓库搬运机器人导航运动控制与NURBS轨迹规划[J]. 机械设计与制造, 2020(7): 290–295
[8] 李峰, 岳云, 张跃进, 等. 基于北斗导航的机器人弹道曲线航道导航系统[J]. 弹道学报, 2021, 33(2): 47–54
