Research On Key Technologies Of Wall Cleaning Robot In High-altitude Complex Environments

In recent years,glass curtain walls have been widely used in high-rise buildings in cities due to their superior lighting performance.However,there are some problems in the cleaning process.The traditional wall cleaning method represented by a worker is not only costly,but also has a high risk factor.Therefore,it is of great practical significance to design and develop a practical and efficient intelligent wall cleaning robot.In order to realize movement and operation of the wall cleaning robot in a complex environment,key structural design,obstacle identification,and path planning issues are analyzed in this paper.The specific research content is as follows:Firstly,based on the actual working conditions of wall cleaning,a polygon track robot with the propeller mixed negative pressure suction device is designed,and the robotics dynamic performance and negative pressure performance were simulated and analyzed to lay the foundation for building a prototype.By establishing a robot dynamics model,and the kinematics and dynamics of the two failure modes of overturning and sliding in the key stage of crossing obstacles are analyzed,the parameters of track and the theoretical driving torque are determined.In view of the stability of the negative pressure adsorption device,the CFD simulation software was used to simulate and analyze the pressure difference inside the cavity during the negative pressure adsorption process.Based on this,experimental prototypes are developed for climbing and obstacle-crossing tests.The results show that the designed robot can stably adsorb on the wall and has good obstacle-crossing performance.Secondly,a robot control system is designed according to the structure and environmental characteristics of the robot,and the functional requirements and control system requirements of the robot are analyzed to further formulate the structure of the robot system.Aiming at the problem of large deviation in the robot's trajectory tracking control process.In order to ensure that the robot can move along a specified path,a dynamic equation of the robotics system is established,and a Simulink simulation model is built to study the trajectory performance of the robot and lay the foundation for robot path planning.Then,research on obstacle detection and recognition methods based on cloud theory.The accurate obstacle recognition is very important for automatic cleaning of wall cleaning robots.Combined with the working principle of the robot and the environmental characteristic,obstacle recognition based on cloud theory ispresented.Based on obstacles wavelet moment,the cloud theory is used to establish the cloud model of the obstacle wavelet moment feature,and the mathematical features in the model are extracted as the support vector machine feature vector for obstacle classification.The results show that three types of obstacles on the wall can be effectively identified by the cloud theory which combined with support vector machine,and the success rate of recognition is over 96%.Finally,based on the research of obstacle recognition and robot trajectory tracking,by establishing a environment map,combining the robot's reciprocating operation method,using the grid method and the mobile robot's global path planning algorithm to establish a global map path planning.The A-star algorithm is used to solve the problem of path planning when the robot is at a dead point,thereby reducing the repetition rate of the cleaning path,so that the robot can complete the path planning of the environment map with 100% cleaning coverage and 5.7% path repetition rate.