An Iterative Learning PID Controller and its Application to a Terrain Simulator

Abstract. Spray-height control systems are widely employed in agricultural sprayers. These systems ensure that the sprayer’s boom are positioned and maintained at a pre-set height above the land/crop while the chemicals are actively being sprayed. To test their performance, a tractor-trailed sprayer is required to expose the control system to a variety of outdoor terrain situations. This method of outdoor testing suffers from poor repeatability due to differences in testing locations through which the tractor is driven. It is further limited during the long winter period. There are also substantial costs associated with travel and transportation of the sprayer apparatus to outdoor testing locations. A promising solution is the use of an indoor terrain simulator. The simulator mainly consists of a platform on which an agricultural sprayer can be loaded. The platform can be controlled to rotate along a horizontal axis and simulate the terrain profiles that would be encountered in the fields during outdoor testing. However, the implementation of motion control on the terrain simulator is challenging due to the system dynamics, which vary depending on the size of sprayer loaded on the platform as well as the temperature and age of the platform-actuated system. In this paper, an adaptive iterative learning proportional-integral-derivative (ILPID) controller is developed to compensate for the time varying system dynamics. A prototype is built, on which the proposed controller is implemented. The experimental results show that the controller parameters can be adaptively adjusted by ILPID and the dynamics of the prototype is better compensated.