Optimizing turning logic of glass curtain wall cleaning robot

Abstract. In contemporary urban environments, there has been a notable increase in the construction of high-rise edifices with glass curtain walls. While glass curtain walls contribute to the aesthetic appeal of buildings, they also give rise to a number of challenges. In recent decades, cleaning robots have been developed to address the issue of maintaining glass curtain walls. However, the challenge of efficiently avoiding obstacles and gaps on the glass curtain remains. The research aims to optimize the trajectory of glass curtain wall cleaning robots. A series of simulations were conducted using MATLAB 2024b and the RobotCraft Robotics Playgrounds add-on. The robot is required to traverse a defined maze, which features a 90-degree turn. The time required for each operational logic to traverse the entire route from the initial point to the final destination has been meticulously documented. The results of these simulations are compared in a chart in order to identify a logic that will result in a faster average speed for the robot. The initial experiment yielded results indicating that the turning is excessively sharp and rigid. Accordingly, the distance detection has been increased in the subsequent iteration to facilitate an earlier turning point. Moreover, an increase in the rotational speed of the motor on the corresponding side was implemented to expand the turning radius, thereby facilitating a more seamless and fluid rotation. In essence, the robot has been optimized to achieve a balance between spatial utilization and high average speed. The refinement of the turning path has been shown to enhance the overall cleaning efficiency while maintaining an exceptional ability to avoid obstacles.