多目标NSDE算法的施釉机器人轨迹优化  

Multi-objective NSDE algorithm for glazing robot trajectory optimization

作　　者：霍平 李伟 陈江峰 冯永利 HUO Ping;LI Wei;CHEN Jiangfeng;FENG Yongli(College of Mechanical Engineering,North China University of Technology,Tangshan 063210,China;Hebei Provincial Industrial Robotics Research Institute,Tangshan 063005,China)

机构地区：[1]华北理工大学机械工程学院,河北唐山063210 [2]河北省工业机器人产业技术研究院,河北唐山063005

出　　处：《重庆理工大学学报(自然科学)》2025年第2期106-112,共7页Journal of Chongqing University of Technology:Natural Science

基　　金：河北省高等学校科学研究计划科技重点项目(ZD2020151);唐山市科技创新团队培养计划项目(21130208D)。

摘　　要：针对机器人施釉中人工示教给定的初始轨迹优化问题,提出一种基于NSDE算法的多目标优化方法。在给定初始轨迹基础上,以釉层厚度均匀性高和施釉工作时间短为优化目标建立多目标优化模型,采用NSDE算法对喷枪移动速度和喷枪高度参数进行优化,计算得优化参数解集,通过仿真分析比较优化前、后釉层厚度的均匀性,并采用优化参数下的喷涂轨迹对坯体试样进行施釉试验及厚度检测,依据此轨迹优化方法喷涂的釉层厚度均匀性误差为9.21%,验证了所规划施釉轨迹的可行性及有效性。We propose a multi-objective optimization method based on the NSDE(non-dominated sorting differential evolution)algorithm for the optimization of the initial trajectory given by manual teaching in robot glazing.On the basis of the given initial trajectory,a multi-objective optimization model is established with the optimization objectives of high glaze thickness uniformity and short glazing time,and the NSDE algorithm is adopted to optimize the parameters of gun moving speed and gun height.The solution set of the optimization parameters is computed,and the uniformity of glaze thickness before and after optimization is compared with the simulation analysis.The glaze test and thickness detection are conducted for the billet specimens using the spray trajectory under the optimization parameters.After measurement,the uniformity error of the glaze layer thickness sprayed using this trajectory optimization method is 9.21%,which verifies the feasibility and effectiveness of our planned glazing trajectory.

关 键 词：施釉机器人 多目标优化 NSDE算法 轨迹优化 厚度均匀性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]