基于半实物仿真的采摘机器人视觉定位研究  

Research on Picking Robot Vision Localization Based on Semi-physical Simulation

作　　者：陈科尹[1,2] 邹湘军[3] 彭红星[3] 梁海英[4] 胡元闯[4] 

机构地区：[1]贺州学院信息与通信工程学院,广西贺州542899 [2]农业部南京农业机械化研究所,江苏南京210014 [3]华南农业大学南方农业机械与装备关键技术省部共建教育部重点实验室,广东广州510642 [4]贺州学院数学与计算机学院,广西贺州542899

出　　处：《系统仿真学报》2017年第8期1736-1746,1761,共12页Journal of System Simulation

基　　金：国家自然科学基金(31571568);广西自然科学青年基金(2015GXNSFBA139264);广西高校科学技术研究(KY2015YB304)

摘　　要：针对基于实物采摘机器人的视觉定位的准确度、稳定性以及试验验证易出错、周期长、不易于室内进行等问题,结合采摘机器人视觉定位机理和机器人运动学,提出了一种基于半实物仿真技术的采摘机器人视觉定位研究方法。该方法采用虚拟采摘机器人代替实物机器人,在虚拟仿真环境下对水果目标采摘点进行定位研究,即视觉定位半实物仿真。试验结果表明:在视觉定位半实物仿真环境下,水果目标定位的平均绝对误差和相对误差分别仅约为2.89 mm和0.43%,满足实际视觉定位精度的要求,从而验证了该仿真方法的有效性和可行性。Aiming at the problems of accuracy and stability of the vision localization based on physical picking robot, which were easy to make mistakes, long cycle, and not easy to be carried out indoors, a picking robot vision localization method based on the semi-physical simulation technology was proposed combining with the robot vision localization mechanism and robot kinematics. This method adopted the virtual picking robot to instead of the physical robot, and studied the vision localization of the fruit target in the virtual simulated environment, which was as the vision localization based on semi-physical simulation. The test results show that： the average absolute error and relative error of the target fruit are only about 2.89 mm and 0.43% in the vision localization simulated environment, which is in line with the accuracy requirements of the actual vision localization, so the effectiveness and feasibility of this simulation method are validated.

关 键 词：采摘机器人 视觉定位 机器人运动学 半实物仿真 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]