Trajectory optimization of wall-building robots using response surface and non-dominated sorting genetic algorithm III

Abstract

Traditional wall-building robots regard brick masonry as a simple assembly process, ignoring the viscoelastic effect of cement mortar, which leads to poor masonry quality. Therefore, this paper proposes a many-objective trajectory optimization method based on response surface methodology (RSM) and non-dominated sorting genetic algorithm III (NSGA-III). Firstly, a substitution model between the objective functions and the design variables is established using RSM, which solves the problem of difficult construction of cement mortar viscoelastic model. Then, a trajectory optimization model is constructed based on seven times non-uniform B-spline, and it is solved using NSGA-III. Finally, the optimal solution is obtained from the Pareto solution set, and it is compared and analyzed with the standard gate-type masonry solution. The results show that after the trajectory optimization, the masonry error is reduced from 2.62 mm to 0.11 mm, which significantly improves the masonry quality, and other performance indicators are also improved. These results contribute to the construction of walls and promote the development of the smart construction industry.