REAL TIME SIMULATOR FOR FIELD SURVEY MOBILE ROBOT

Abstract

A systematizing elucidation to the hazards present in the current human health due to spraying of conceivable toxic chemicals in the detained space of an agricultural field or hot and steamy glasshouse is accomplished by the design and disposition of a sovereign mobile robot for use in disease prevention and pest control applications in monetary greenhouses. An Embedded system based stereotyped robot is designed for this purpose. The platforms ability to fortunately head out for itself down the platform of a greenhouse shows the forcefulness of this platform, while the pesticide sprinkling system is used to efficiently spray the plants evenly with set dosages. A vision based robotic control regulation strategy is developed for a non-calibrated camera system which is mounted on a wheeled single mobile robot contingent on non-holonomic motion restraint, which can drive the mobile robot to the target position with exponential convergence. Subsequently, by using the retrieved camera intrinsic parameters, a straight-line motion controller is developed to drive the robot to the desired position, with the coordination of the robot always facing the target position. By the proposed method, the robot can be confined in map-free and GPS-free environments, and the results of localization can be theoretically proved confluent to their real values and thriving to the measurement noises. The performance of the proposed method is further validated by both simulation and experimental results.