Design And Test Of Control System For Selective Harvesting Of White Asparagus Based On Machine Vision

Asparagus is a perennial herbaceous plant,which contains more than five times more essential protein,vitamins,amino acids and carbohydrates than common vegetables.It has the reputation of “King of Vegetable”.It is also known as “White Gold” because of its rich nutrition,white as jade.At present,asparagus industry in the world is in the stage of transferring from developed countries to developing countries,and China has become the main undertaking place.After more than 20 years of development,China's asparagus industry has developed into a new industry with competitive advantages.The production of asparagus ranks first in the world in terms of area and output.At the same time,with the development of economy and society,China has become an increasingly important asparagus consumer market.For a long time,the harvesting of white asparagus in domestic and foreign has been mainly manual,with high labor intensity,high labor cost and low scale.Therefore,the development of white asparagus selective harvesting robot is of great significance to improve the scale and industrialization of asparagus industry in China,to enhance the level of Intelligent Agriculture in China and to realize agricultural modernization.In this paper,a control system for selective harvesting robot of white asparagus was designed for harvesting problem of white asparagus in mature period.Research contents of this project are as follows:Firstly,according to the system requirements,the control system of the selective asparagus harvesting robot is divided into two parts: upper and lower computer.The upper computer hardware system mainly includes machine vision system and main control module,mainly realizes the function of image collection and transaction.Lower computer hardware system mainly includes STM32 controller,harvesting execution subsystem and self-propelled subsystem,which mainly realizes the function of white asparagus harvesting and the overall movement of the robot.By analyzing the visual requirements in the process of harvesting,a suitable machine vision system is built and related hardware are selected.Through the functional analysis of lower computer system,select the key hardware of the lower computer control system and study its working principle.Secondly,the environment of white asparagus in the natural growth environment is very complicated,and the complex soil background and the small area of the target area have undoubtedly brought great difficulties to recognition and positioning of machine vision.In this study,the camera is first calibrated and the acquired image is corrected to meet the accuracy requirements of post-image processing.With the help of image processing technology,a threshold segmentation image processing algorithm based on RGB three components is proposed to preliminarily process the soil background and interferents.A contour information-based processing algorithm is proposed.The contour is rounded and area-detected,and the region of interest is extracted and texture feature detection based on the gray level co-occurrence matrix is performed to determine the contour of white asparagus.In the contour area,the contour center point extraction is performed to obtain pixel coordinate information corresponding to white asparagus tip.Finally,according to the idea of modularization and the functional requirements of the system,the software design and joint debugging of upper computer control system based on industrial PC and lower computer main control system based on STM32 are carried out respectively.A coordinate transformation model is established to obtain the moving distance of the end-effectors of the harvesting robot in the world coordinate system corresponding to the target coordinate points in the pixel coordinate system.A path optimization algorithm is proposed for the harvesting process of white asparagus.A synchronous control technique is proposed for the end-effectors positioning process.According to the communication requirements between upper and lower computer systems,a communication cooperation based on RS232 is developed.The software and hardware of the selective asparagus harvesting control system were debugged and the debugging results were analyzed.A large number of harvesting tests were carried out to obtain the recognition accuracy and harvesting positioning accuracy of the white asparagus.The source of error was analyzed,and the feasibility of the control system for selective harvesting of white asparagus robot was verified.