David Ooms

Measurements of the horizontal sprayer boom movements by sensor data fusion

A study investigated the horizontal movements of trailed sprayer booms with the aim of distinguishing their yaw and jolt motions as well as their deformations. Two vehicles were equipped with a radar speed sensor and a three-axis dynamic measurement unit, while each boom was instrumented with ultrasonic sensors and accelerometers. Sensor fusion was used to extract yaw, jolt and deformation speeds by assuming that the motion of any point of the boom is equal to the superimposition of complementary high and low-frequency motions, delimited by a cut-off frequency, estimated on the basis of laboratory trials. Tests performed on trailed sprayers equipped with 22 and 24 m boom lengths, on different soils, indicated that yaw, jolting and deformation, respectively, occurred near 0.3, 2 and 1 Hz. Velocity variations relative to the forward speed of the vehicle were lower than 0.57 m/s and were reached at a speed of 3.5 m/s (12.7 km/h). Ground deposit, roughly estimated as proportional to the inverse of the speed, exhibited coefficients of variation between 4 and 6% for all tests.

Impact of the horizontal movements of a sprayer boom on the longitudinal spray distribution in field conditions

Abstract The influence of horizontal sprayer boom movements on the longitudinal distribution of the sprayed product was investigated in field and laboratory conditions. The boom movements of a sprayer were measured in the field using a specially designed measurement chain, and the resulting ground distribution was estimated by spraying a dye (Nigrosine) on large sheets of wallpaper and using image analysis to compute the spray coverage. The horizontal movements, identical to those measured in the field, were applied to a laboratory boom part moving on horizontal rails, in the absence of wind. For the three tested situations, the longitudinal coefficients of variation (CV) of the spray coverage were very close in the field (8.81%, 6.9%, 10.9%) and in the laboratory (9.17%, 8.17%, 10.96%). This indicates that horizontal movements were the main source of spray deposit variations, even if other factors such as vertical movements or wind may have an influence. The mean ratio RCV between the CV of the horizontal movements and the spray coverage was estimated. In addition, the amplitude of the boom movements was investigated in a large range of situations (five different sprayers, three different crops). On basis of the measured movements and the computed RCV, the variability of the longitudinal spray distribution due to horizontal movements was estimated at between 3.5% and 5.2%.

A sorting optimization curve with quality and yield requirements

Binary classifiers used for sorting can be compared and optimized using receiver-operating characteristic (ROC) curves which describe the trade-off between the false positive rate and true positive rate of the classifiers. This approach is well suited for the diagnosis of human diseases where individual costs of misclassification are of great concern. While it can be applied to the sorting of merchandise or other materials, the variables described by the ROC curve and its existing alternatives are less relevant for that range of applications and another approach is needed. In this paper, quality and yield factors are introduced into a sorting optimization curve (SOC) for the choice of the operating point of the classifier, associated with the prediction of output quantity and quality. Given examples are the sorting of seeds and apples with specific requirements. In both cases the operating point of the classifier is easily chosen on the SOC, while the output characteristics of the sorted product are accurately predicted.