Review of AGV Obstacle Avoidance Algorithm Based on Machine Vision

Abstract

With the development trend of unmanned and intelligent, it has become a development trend for AGV to leave human operators and perform diversified tasks. However, operating in complex environments such as urban streets, mines, and construction sites is a difficult problem that needs to be solved currently. Obstacles in complex environments pose a serious threat to the AGV trolley in operation. Obstacle avoidance technology has become a key part of the AGV trolley’s task decision-making system and plays an important role in ensuring the safe operation of the AGV trolley and improving work efficiency. This article first elaborates on the concept of the AGV trolley obstacle avoidance algorithm and the evaluation criteria for the optimal path, then elaborates and compares the advantages and disadvantages of different types of obstacle avoidance algorithms such as optimization-based, potential field-based, and machine learning-based obstacle avoidance algorithms, and finally obtains the research focus and direction of the AGV trolley.