Development of topographic maps and rear steering control for an agricultural vehicle through incorporation of posture and attitude measurements
The two articles contained in this work investigate the relationship between an agricultural vehicle's posture and attitude with (a) its surroundings and (b) machine performance. In the first case, a self-propelled agricultural sprayer was equipped with four RTK DGPS receivers and an inertial measurement unit (IMU) to measure vehicle attitude and field elevation as the vehicle was driven across a field. Data was collected in a stop-and-go fashion as well as at three different speeds on a field area with varying topography. Using ordinary kriging, digital elevation models (DEMs) were interpolated from the elevation measurements and elevation plus attitude measurements. The resulting DEMs were compared to each other to evaluate the effect of including attitude measurement on DEM accuracy. At the widest swath width, the DEMs generated with attitude measurements had substantially lower error measures than those DEMs generated without attitude measurements. These results provide evidence that support the feasibility of using vehicle-based measurements collected during typical field operations for accurate DEM development. In the second case, a steering controller was designed and implemented on a self-propelled agricultural sprayer with four-wheel steering (4WS). The goals of this controller were to reduce the off-tracking error of the rear wheels and control turning radius during lateral shifts to reduce chemical application error. The vehicle was driven along marked courses of different shapes to test the steering controller's performance. A computer simulation provided an estimate of chemical application rates across the spray boom during lateral shift maneuvers. During hillside operations, the controller was able to reduce the area damaged by the rear wheels from 107.35 m2 using two-wheel steer (2WS) to 0.32 m2 with Active Rear Steering (ARS) control. During 90-degree turns, the controller reduced the area damaged by rear wheels from 49.34 m2 in 2 WS to 1.15 m2 with ARS. This reduction in rear wheel off-tracking could lead to a reduction in crop damage through turns and during hillside operation, as well as reduced chemical application errors during turns.