Font Size: a A A

Research On Stability Control And Location Sensing Method For Glass Curtain Wall Cleaning Robot

Posted on:2021-02-11Degree:MasterType:Thesis
Country:ChinaCandidate:B FengFull Text:PDF
GTID:2392330611998052Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
In modern cities,more and more glass curtain walls of skyscrapers need to be cleaned.For the efficiency and safety,robotic automation is well suited to this work scenario.However,the main cleaning method is still manual cleaning.The task of cleaning glass curtain wall is faced high labor intensity,labor cost and risk.In response to the problems of poor adsorption effect and low obstacle crossing ability of the existing cleaning robot,a cleaning robot based on flexible rope drive is designed.Four ducted fans arranged in the cross shape are adopted to ensure the adsorption performance and the ability to cross obstacles.The tilting mechanism of rotating fans is designed to change the thrust angle in order to resist cross wind and improve safety.Some strategies such as spray pretreatment are used to improve the cleaning function and enlarge the horizontal cleaning area.In order to ensure the motion stability of the robot when it is working,especially crossing obstacles,the stability control method is studied.For attitude angle are susceptible to interference problems,we analyze four ducted fans based on cross layout attitude angle control method which can decouple the control of pitching angle and roll angle,then a dynamic model of robot body is established.For pitching angle channel,roll angle and fly height channel,the controller is designed,respectively.The simulation verifies the effect of the controller.In order to realize robot autonomy and improve the accuracy of localiz ation perception,a localization perception method based on sensor fusion is studied.Through the data fusion of vision and inertial sensors,the IMU information between frames was pre-integrated and aligned with the image information,and the nonlinear objective optimization function was established to realize the pose estimation algorithm.The high-precision attitude angle can be obtained by the pose estimation algorithm.Aiming at the estimation of robot position coordinates,a coordinate estimation equation is established based on multi-sensor.In order to verify the effectiveness of the design scheme and simulation analysis,the prototype is made.The program is completed.The experiments verify the stability of the prototype and the effectiveness of the measurement and control system.
Keywords/Search Tags:glass-curtain-wall cleaning, robot design, stability control, sensor fusion, location sense
PDF Full Text Request
Related items