Abstract
Intelligent multi-robot systems get more and more deployed in industrial settings to solve complex and repetitive tasks. Due to safety and economic reasons they need to operate dependably. To ensure a high degree of dependability, testing the deployed system has to be done in a rigorous way. Advanced multi-robot systems show a rich set of complex behaviors. Thus, these systems are difficult to test manually. Moreover, the space of potential environments and tasks for such systems is enormous. Therefore, methods that are able to explore this space in a structured way are needed. One way to address these issues is through model-based testing. In this paper we present an approach for testing the navigation system of a fleet of industrial transport robots. We show how all potential environments and navigation behaviors as well as requirements and restrictions can be represented in a formal constraint-based model. Moreover, we present the concept of coverage criteria in order to handle the potentially infinite space of test cases. Finally, we show how test cases can be derived from this model in an efficient way. In order to show the feasibility of the proposed approach we present an empirical evaluation of a prototype implementation using a real industrial use case.
Original language | English |
---|---|
Title of host publication | 2019 IEEE International Conference On Artificial Intelligence Testing (AITest) |
Pages | 129-137 |
DOIs | |
Publication status | Published - 2019 |
Event | IEEE International Conference On Artificial Intelligence Testing: AITest 2019 - Doubletree by Hilton Newark - Fremont, Fremont, United States Duration: 4 Apr 2019 → 9 Apr 2019 |
Conference
Conference | IEEE International Conference On Artificial Intelligence Testing |
---|---|
Abbreviated title | AITest |
Country/Territory | United States |
City | Fremont |
Period | 4/04/19 → 9/04/19 |
Fields of Expertise
- Information, Communication & Computing