基于多尺度目标物特征分析的自由度机器人抓取控制系统  

Degree-of-freedom robot grasping control system based on multi-scale target feature analysis

作　　者：曹桂林 CAO Guilin(College of Information Engineering,Guangzhou Vocational and Technical University of Science and Technology,Guangzhou 510080,China)

机构地区：[1]广州科技职业技术大学信息工程学院,广州510080

出　　处：《自动化与仪器仪表》2024年第12期173-177,共5页Automation & Instrumentation

摘　　要：出于解决自由度机器人在动态不确定的复杂工作环境中抓取任务准确性低、抗干扰性较差等问题的目的,提出一种基于多尺度目标物特征分析的自由度机器人抓取控制系统;自由度机器人抓取控制系统旨在提高复杂环境下自由度机器人抓取控制的准确性与实时性,同时增强自动化抓取系统面对环境干扰时的稳定性;系统通过金字塔特征分析以及深度学习模型优化自由度机器人自动化抓取策略,在初步确定抓取区域之后细化机器人抓取姿态;设计关于自动化抓取准确性、稳定性以及及时性的对比实验,实验结果表明基于人工智能技术的自由度机器人自动化抓取的准确率为70%至93%,同时抗干扰性以及实时性均优于其他算法。In order to solve the problems of low accuracy and poor anti-interference of the grasping task of the degree-of-freedom robot in the dynamic and uncertain complex working environment,a degree-of-freedom robot grasping control system based on multi-scale target feature analysis was proposed.The DOF robot gripping control system aims to improve the accuracy and real-time performance of DOF robot gripping control in complex environments,and at the same time enhance the stability of the automated gripping system in the face of environmental interference.Through pyramid feature analysis and deep learning model,the system optimizes the automatic grasping strategy of the DOF robot,and refines the grasping posture of the robot after the initial determination of the grasping area.Comparative experiments on the accuracy,stability and timeliness of automatic grasping are designed,and the experimental results show that the accuracy of automatic grasping of robots based on artificial intelligence technology is 70%to 93%,and the anti-interference and timeliness are better than other algorithms.

关 键 词：人工智能技术 深度图像分析 金字塔特征分析 ROI池化 回归损失 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置] TP273[自动化与计算机技术—控制科学与工程]