Multi-robot Task Assignment Method in the Construction Waste Sorting System

Abstract

In an automated construction waste sorting system, multiple waste collection robots working in parallel are responsible for grabbing the waste on the conveyor belt and putting it into the waste container. In order to reduce the occurrence of missed picking and increase the speed of waste collection, it is necessary to assign the collection tasks of multiple robots appropriately. In this paper, a novel multi-robot task assignment method based on the multiple traveling salesman problem (MTSP) is proposed, which ingeniously transforms the multi-robot task assignment problem for construction waste collection tasks into a MTSP, and the genetic algorithm (GA) is used to solve it. We take the case of two robots collecting 50 pieces of waste as an example for simulation. The results show that this method can effectively reduce the number of missed pickings and increase the speed of waste collection. At the same time, the algorithm has good convergence.