基于双环滑模控制的全向移动工业机器人多目标点跟踪控制系统设计
Design of Multi target Point Tracking Control System for Omnidirectional Mobile Industrial Robot Based on Double Loop Sliding Mode Control
周峻水1
作者信息
- 1. 北京奔驰汽车有限公司,北京 100176
- 折叠
摘要
全向移动工业机器人在实际应用场景中,经常处于存在较高噪音的环境中,导致多目标跟踪的性能变差,为此设计基于双环滑模控制的全向移动工业机器人多目标点跟踪控制系统;系统硬件设计了主控制器模块及主控制器芯片;在数据采集模块中配置数据采集卡,实施帧数据采集;在无线通信模块中,采用Wi-Fi无线通信技术,实现系统的远程控制、数据传输功能;系统软件部分设计基于GhostNet的改进目标检测网络,实施多目标点检测;设计基于改进SORT算法的多目标点跟踪算法,实现多目标点跟踪;设计基于双环滑模控制的多目标点跟踪控制器,实现多目标点跟踪控制;测试结果表明,静态障碍物情况下,在仓库场景下,设计系统的目标点跟踪准确性平均值为97.23％,室外场景下的目标点跟踪准确性平均值为96.37％;动态障碍物情况下,两种场景下系统的目标点跟踪准确性略低于静态障碍物情况.
Abstract
In practical application scenarios,omnidirectional mobile industrial robots often work in high noise environments,re-sulting in poor performance of multi-objective tracking.Therefore,a multi-objective point tracking control system for omnidirectional mobile industrial robots based on double loop sliding mode control is designed.Design the system's main controller module and main controller chip;Configure a data acquisition card in the data acquisition module to implement the frame data acquisition;In the wire-less communication module,Wi-Fi wireless communication technology is used to achieve the system's remote control and data trans-mission functions.The system software design adopts an improved object detection network based on GhostNet,implementing multi object point detection.the multi-objective point tracking algorithm based on the improved SORT algorithm is designed to achieve the multi-objective point tracking.The multi-objective point tracking controller based on dual loop sliding mode control is designed to a-chieve the multi-objective point tracking control.Experimental results show that in static obstacles,the average target tracking accu-racy of the designed system in warehouse scenarios is 97.23％,and the average target tracking accuracy of outdoor scenarios is 96.37％;The system target tracking accuracy of dynamic obstacles is slightly lower than that of static obstacles.
关键词
双环滑模控制/全向移动工业机器人/多目标点跟踪/主控制器Key words
double loop sliding mode control/omnidirectional mobile industrial robots/multi target tracking/main controller引用本文复制引用
出版年
2024
NETL
NSTL
万方数据