Implementing a Time Optimal Task Sequence For Robot Assembly Using Constraint Programming

Abstract

Assembly lines in a lean production environment may persist for months or years, but due to the rapid change in demands on the consumer market they can be subject to quick changes. Robots have been proved to handle tasks that previously were limited to human workers: this is the goal of the flexible two armed robot developed by ABB. When installing a new robot into an assembly line there are several parameters which make it a difficult job for programmers, even experienced ones, to install the robot. These problems lead to long installation processes that can take weeks, and there are great benefits of automating the process of finding good solutions to the problem. In this thesis a constraint programming approach is presented as a way to solve the complex sequencing problem when installing a two armed robot into a new environment. When benchmarked against a reference case study, the implemented prototype solutions showed an improvement of 17%, all within a time limit of 20 minutes instead of weeks. This shows that constraint programming can be a good tool for automating robot installations.