Experimental and theoretical analysis on truss construction robot: Automatic grasping and hoisting of concrete composite floor slab

Abstract

AbstractThe development of various construction robots can significantly reduce the construction labor intensity and risk. We present an experimental task of a truss construction robot, including automatic grasping and hoist positioning of concrete composite floor slab. And, a control strategy matching the construction robot is proposed to solve the following problems: (a) Due to the widespread manufacturing errors of building components, the position deviation of diagonal web reinforcement on concrete composite floor slab may cause grasping failure. (b) The chaotic surface of the floor slab seriously affects the recognition accuracy of the grasping target. (c) The large mass and inertia of heavy‐duty hanger and concrete composite floor slab make it difficult for robot to achieve accurate positioning under eccentric load condition. The machine vision recognition system ensures that if the diagonal web reinforcement is detected more than half of the grasping path, the grapple hook can change the grasping path independently to prevent it from colliding with the offset diagonal web reinforcement. An anticollision and grasping path planning method based on machine vision is proposed to improve the success rate and efficiency of automatic grasping. The mean value of proportional speed and proportional position is introduced as the evaluation speed and evaluation position of driving wheel to improve the accuracy of multiwheel synchronization. Moreover, we analyzed the robot performance from actual grasping and hoist positioning process of the concrete composite floor slab. The experimental results show that the robot has a good performance on the automatic grasping and hoisting of the concrete composite floor slab.