A research on the development of GPS guided an agricultural robot and use the sample of stubble density determination

In this study,mobile robot that can move on the field was designed and conducted.The mobile robot is a four-wheel vehicle which steered with two DC motors.GPS system was used for steering and driving autonomously for mobile robot.The navigation software was developed for processing data,which came from the GPS system.The mobile robot?s drives and turns were achieved by the same or different DC motors? speed.The control signals for the dc motors were sent by the panel computer which was placed on the vehicle.Azimuth and heading angles are used to guide the vehicle to the desired point.For the driving of the vehicle to the target point,distance between the current point of the mobile vehicle and the target point is calculated.By the help of developed navigation software,heading angle,azimuth angle and the distance between the current and target point were continuously calculated.When the difference between the heading and the azimuth angle,and the distance becomes closer to zero,mobile vehicle is reached to target point.In this study,for the determination of the stubble density,image processing software has been developed with the aim of adding functionality to the mobile vehicle.The stubble images in the field have been obtained by the using digital camera which was mounted mobile vehicle.The colored stubble images have been converted to grayscale images by the using developed software.In this way,the stubble color value was determined.Thus,the intensity of stubble color value in the grayscale image was counted and so,the ratio of the stubble was determined by the software.Field experiments were conducted in two different fields in July and August 2012.30 different GPS point were used for each trial.Each experiment was carried out 3 replicates for 10 points.The distance between mobil robot?s stop point and target point was measured.So,target point precision of the mobil robot was determined.In addition,the stubble density ratios were determined for each point by the method of image processing.The stubble density ratios which were determined in tests by the image processing were compared with the line transects method.To evaluate the test results,independent sample t test and arithmetic mean was conducted.In the results of the experiments,it was founded that the lineer target point precision ranged from 10 to 12 cm and the distributed target point precision ranged from 15 to 17 cm In the process of determining the stubble density,differences between the results of both methods were not found in the first experiment (P = 0.193).In the second experiment,similarly,differences between the results of both methods were not found (P=0.578).KEY WORDS: Mobile Robot, GPS, Navigation, Image Processing, Stubble Density