The Study And Development Of Multiple AGV Coordinated Control System

In the times of the globalization of industry 4.0,the prologue to the ‘Made in China 2025' has been open.Logistics Robot which occupies the important position in the global market for robots industry has broad prospects.The intelligent logistics system is developing rapidly which can save a lot of labor costs.Logistics system is developing toward the intelligence,integration,hierarchy,flexibility and socialization.The purpose of the paper is to study the coordinated control problem of multiple AGV(Automated Guided Vehicle)under leader-follower framework.AGV is the key equipment of logistics system because of its reliability,flexibility,adaptability and so on.So this research has important theoretical significance and higher practical application value.This paper mainly includes the following two aspects.Firstly,we focus on the study of intelligent AGV,this paper develops the location system,obstacle-avoidance and protection device of omni-directional mobile AGV with Mecanum Wheel.The indoor positioning system based on laser is constructed independently in order to provide precise navigation location information for leader AGV.The relative position of follower to leader AGV is measured by ultrasonic positioning and following system.In order to ensure AGV running in narrow space,the paper proposes the image recognition feature points algorithm for indoor location and posture of AGV.The obstacles detection and protection device is designed for improving the safety of AGV operation.Secondly,we focus on the study of multiple AGV coordinated control system.The experiment platform for multiple AGV control system based on C# is developed,which provides basis for the data exchange between PC and AGV.For the consensus problem of leader-follower AGV system with bounded acceleration,a finite-time consensus controller based on non-singular fast terminal sliding mode is designed.The simulation results show that position and velocity of leader is tracked precisely by followers in finite-time,and the algorithm is valid for AGV system.It provides a reliable theoretical support for multiple AGV coordination control.