A control of collision and deadlock avoidance for automated guided vehicles with a fault-tolerance capability

Qin Li, A.Y. Pogromskiy, A.C. Adriaansen, J.T. Udding

Research output: Contribution to journalArticleAcademicpeer-review

37 Citations (Scopus)
173 Downloads (Pure)


Based on a novel discrete-event zone-control model, in our previous papers [1, 2], we presented a time-efficient traffic control for automated guided vehicle (AGV) systems to exclude inter-vehicle collisions and system deadlocks, together with a case study on container terminals. The traffic control allows each vehicle in an AGV system to freely choose its routes for any finite sequence of zone-to-zone transportation tasks and the routes can be constructed in an online fashion. In this paper, we extended our previous results with two practical goals: (1) to increase the utilization of the workspace area by reducing the minimally allowed area of each zone; (2) to avoid vehicle collisions and deadlocks with the occurrence of vehicle breakdowns. To achieve the first goal, we include one extra vehicle event that allows each vehicle to probe further ahead while it is moving on the guide-path. This leads to an extension of our previous discrete-event model and traffic control rules, which are presented in the first part of the paper. The second part of the paper concerns the second goal, for which an emergency traffic control scheme is designed as supplementary to the normal traffic control rules. As in our previous papers, the improved model and traffic control are applied to a simulation of quayside container transshipment at container terminals; our simulation results are compared with those from two interesting works in the literature.
Original languageEnglish
Article number64
Number of pages24
JournalInternational Journal of Advanced Robotic Systems
Issue number2
Publication statusPublished - 7 Apr 2016


Dive into the research topics of 'A control of collision and deadlock avoidance for automated guided vehicles with a fault-tolerance capability'. Together they form a unique fingerprint.

Cite this