| It is significantly important to level paddy field,which can improve crop yields by 20%~30%,reduce water use by 30%~50%,and reduce weeds growth and fertilizer loss.In order to meet the demands of levelling operations and reduce the dependency on laser transmitter,a novel method based on field character istics for measuring leveler height was put forward.A 2D laser range finder(HOKUYO,UTM-30LX)mounted above the leveler was used to measure the distance between the 2D laser ranger finder and the surface of paddy field.The vertical distance needed by leveler control system was finally estimated after attitudes correction of the 2D laser rangefinder with AHRS(attitude and heading reference system,Mti-300).The main research contents are as follows:1.Laser reflection characteristics of muddy water in tilled paddy field.After analyzing the distance measurement principle based on water surface of paddy field,the experimental study on the laser reflection of muddy water with different sedimentation time was carried out.Data acquired and analyzed by the software URG show that: 1)distance measured increased from 352.7mm to 385.7mm in the first 10.5h;2)the distance measured stayed relatively stable between 10.5~24h,which increased merely 0.8mm/h;3)the distance could not be measured correctly if the muddy water was precipitated for too long.Therefore,the proper time for levelling is 10.5~24h after the field has been tilled.If the field had to be levelled within 10.5h,keep the operation time to a minimum,or the water surface based reference will bring a rather big error.2.Water surface based reference recognition algorithm.The Water surface recognition algorithm was put forward to obtain the basis reference.The distance data was clustered by using the principle of mode,and the distance between the 2D laser range finder and the water surface was obtained by weighting fused the clustering data and the empirical distance value of the previous time.The room floor was modeled as the surface of paddy field and several boxes were simulated as other things in field to carry out a test.The result shows that:1)the maximum error was less than 7mm when the 2D laser range finder was mounted in different height(1300~1800mm)to vertically scan water surface;2)the maximum error of boxes surface measurement was less than 11mm;3)proportion of boxes on based reference has influence on measurement accuracy,and if the proportion of the floor was no less than 21.38%,the maximum measurement error was 15 mm 3.Water surface based reference recognition algorithm with attitude correction.The water surface based reference recognition algorithm with attitude correction was put forward to correct the influence of tractor roll and pith on measurement.The distance measured by the 2D laser range finder was corrected by the inclination angle measured by the AHRS.Tests with 7 different inclination angles(-15°~+15°)showed that the absolute error of measurements was no more than 5mm if the range of pitch and roll are limited to ±15°respectively.4.Field experiment.Validity of the water surface based reference recognition algorithm was verified on the 1PJ-3.0 laser leveler machine in paddy field.The results showed that the RMSE of measurement was no more than 6.56 mm,the absolute error of measurement was no more than 10 mm.This water surface based reference recognition algorithm had a higher-resolution and advanced precision measurement when comparing with the traditional laser receiver based way.Error measured by this new algorithm distributes within the range of laser receive,which meets the accuracy demand of leveler height control system. |
PDF Full Text Request