摘要
提出了一种融合力觉与视觉信息的协作机械臂高精度自主作业系统,采用分层式架构设计,核心由多传感器信息融合框架与分布式运动控制模块构成。系统通过六维力/力矩传感器与高分辨率工业相机实现数据采集,采用改进的卡尔曼滤波算法动态分配力觉与视觉权重系数,实现多源信息互补增强。关键技术指标包括±0.02mm的重复定位精度、20ms的全链路响应时间以及5ms内的数据融合延迟。在电力系统应用中,系统通过力觉—视觉融合实现了微米级配合使机械臂自主作业成功率提升至98%以上。
This study proposes a high-precision autonomous operation system for collaborative robotic arms that integrates force sensing and visual information.The system is designed with a layered architecture and consists of a multi-sensor information fusion framework and a distributed motion control module.The system collects data through six axis force/torque sensors and high-resolution industrial cameras,and uses an improved Kalman filtering algorithm to dynamically allocate force and visual weight coefficients,achieving complementary enhancement of multi-source information.The key technical indicators include a repeatability accuracy of±0.02mm,a full link response time of 20ms,and a data fusion delay within 5ms.In power system applications,the system achieves micrometer level coordination and constant force polishing control through force vision fusion,increasing the assembly success rate to over 98%and polishing efficiency by three times compared to manual labor.
作者
任光海
杨成军
冯安住
Ren Guanghai;Yang Chengjun;Feng Anzhu(Shandong Shengde Intelligent Technology Co.,Ltd.,Jinan,Shandong 250000,CHN)
出处
《模具制造》
2025年第11期204-206,共3页
Die & Mould Manufacture
关键词
力觉融合
视觉引导
高精度
电力系统
force perception fusion
visual guidance
high precision
power system
