Toru Osuga’s research while affiliated with National Institute of Information and Communications Technology and other places

Robotic automation is becoming prevalent in the manufacturing industry, improving productivity and saving labor. Mobile robotic systems such as Automated Guided Vehicles (AGVs) play critical roles in transport and storage of parts and products. Maintaining reliable wireless communications between AGVs and their control system is essential for high productivity, but often difficult in physically complex wireless environments of storage system. Specifically, signal loss due to shadowing can trigger operational delays due to the need to check the status of storage operations. On the other, avoiding signal loss requires using more wireless resources. So, it is important to analyze and evaluate the wireless resources required to guarantee efficient operation. However, it is difficult to combine existing simulators for wireless systems and factory systems to analyze inter-dependencies of these systems. This paper presents a method for building a simulation model of a 3D storage system with wireless communication based on multi-layer system analysis. The method is applied to the evaluation of the storage performance and wireless channel usage of a 4-level storage system, including a comparison of the effects of fixed and adaptive rate control. Simulation results shown that adaptive rate control based on learning a spatial map of link quality can achieve reliable storage with up to 50 % reduction in channel load ratio compared to the use of fixed rates. The results demonstrate the usefulness of this type of model for evaluating performance in factory sites with complex wireless environments.