A path generation algorithm of an Automatic Guided Vehicle using sensor scanning method

Abstract

In this paper, a path generation algorithm that uses sensor scannings is described. A scanning algorithm for recognizing the ambient environment of the Automatic Guided Vehicle (AGV) that uses the information from the sensor platform is proposed. An algorithm for computing the real path and obstacle length is developed by using a scanning method that controls rotating of the sensors on the platform. The AGV can recognize the given path by adopting this algorithm. As the AG V with two-wheel drive constitute a nonholonomic system, a linearized kinematic model is applied to the AGV motor control. An optimal controller is designed for tracking the reference path which is generated by recognizing the path pattern. Based on experimental results, the proposed algorithm that uses scanning with a sensor platform employing only a small number of sensors and a low cost controller for the AG V is shown to be adequate for path generation.

A real-time path planning approach without the computation of Cspace obstacles

“…Real-time sensor fusion or computer vision signal processing for environmental map building; 4: Path planning based on the results obtained in step 3; and 5: High level mission management or control. [8][9][10][11][12] In this paper we concentrate on the path planning task for the vehicle through a cluttered environment, such as around a subsea structure. Although seemingly trivial, it has proved notoriously difficult to find techniques which work efficiently in the presence of multiple obstacles.…”