자율주행로봇의 정체 회피를 위한 기계학습 기반의 경로선정

초록

Recently interest in autonomous mobile robots (AMRs) has been increasing in various fields such as manufacturing and service. In particular, the demand for automated material handling in smart factories is gradually increasing. We deal with the problem of path selection to minimize the transportation time of AMR in job shop manufacturing systems that operate multiple AMRs. A machine learning approach for path selection has been proposed to avoid congestion paths. Simulation experiments were conducted to compare the performance of various binary classification and anomaly detection algorithms. The proposed method improved the transport delay ratio. Complementary points and research guidelines for future improvement research are presented in detail.

키워드