Research On Some Key Technologies Of Indoor Intelligent Service Robot

As a high-tech intelligent product,robot integrates many subjects and fields such as mechanical design,control technology,electronic technology,communication technology,computer technology,sensor technology and so on.With the increasing social problems such as aging and life pressure increasing,service robots are gradually being applied to people's daily life,so they have a promising application in market.However,the research for service robot is in the early stage,the high design cost and some technical problems have not been solved very well,so this paper focuses on the key technology of service robot with high security and low cost to do my study,the main research contents are as follows.First of all,the background and significance of indoor intelligent service robot are introduced.The research situation of service robot at home and abroad is summarized,and the key technology of indoor service robot is analyzed.A low-cost indoor service robot research platform is designed,including robot body structure,power components,and pneumatic arm.The design of the remote control handle system,the touch screen system,and the voice system has been completed.Secondly,we have mastered the fabrication method of pneumatic artificial muscles and made the pneumatic artificial muscle arm.The static model of pneumatic artificial muscle is analyzed,and the output displacement performance is tested based on the performance test platform.According to the experimental results,the hysteretic model of pneumatic muscle is established by using EUPI.The drive control system of pneumatic arm is designed based on Arduino controller to realize the motion of pneumatic muscle arm.Then,the robot vision system based on Astra camera is designed.The internal and external parameters of Astra camera were calibrated by using camera calibration and MATLAB calibration toolbox,so that the transformation relationship between the coordinate systems of the vision system is determined,and the registration of color cameras and deep cameras is realized.The characteristics of several image recognition networks are analyzed,and an object recognition and detection experiment is done based on Faster R-CNN recognition framework.Finally,using laser radar,encoder and MPU6050 as sensing elements,robot autonomous navigation system is designed based on ROS,and robot relevant model transformation relationship is determined.The functions of robot localization,composition,navigation and obstacle avoidance are realized.The robot navigation system is proved with good performance by experiments.The whole experiment test of robot is carried out in a simple experiment scene.