Research On Autonomous Driving System Of Small Plant Protection Vehicle In High Pole Environment Based On Slam Technology

With the continuous development of digital technology and point cloud processing technology,this technology also has many applications in small agricultural plant protection vehicles.At present,more and more scientific research institutions and enterprises are improving the intelligent level of agricultural small-scale plant protection vehicles,such as the navigation,obstacle avoidance and identification of intelligent equipment and other precision operations has become a hot research topic.Therefore,the point cloud processing technology plays an important role in the intelligent agricultural machinery equipment,and in the point cloud processing technology,it needs sensors to accurately obtain the agricultural environment and complete the positioning function of crops in the agricultural environment information,which has a great challenge to the development of intelligence.This paper is based on the slam technology of small-scale plant protection vehicle autonomous driving system in high pole environment,which lays the foundation for small-scale agricultural plant protection vehicle to drive in the field and identify the passable area.Point cloud processing technology first uses three-dimensional radar sensor to obtain three-dimensional crop information in the field.Combined with the existing point cloud processing technology,an algorithm is proposed to accurately identify the field crop position coordinates,identify the field accessible area of agricultural small-scale plant protection vehicles in the three-dimensional environment,and provide the basis for the navigation of agricultural small-scale plant protection vehicles in the field.The specific research is as follows:First of all,according to the demand,this paper designs a platform of agricultural plant protection vehicle,which is driven by rear motor,controlled by STM32 microprocessor,and controlled by front wheel motor.Then,based on the three-dimensional lidar sensor,industrial computer and agricultural small plant protection vehicle platform,a three-dimensional point cloud collection system in the field is built.Using slam technology to collect the point cloud data of corn plant model in the field,and then through the industrial computer to process the point cloud,filter the irrelevant point cloud data,and then carry out the subsequent crop positioning processing,and independently run the agricultural small plant protection vehicle Drive for control.Secondly,the three-dimensional crop point cloud is identified by clustering segmentation algorithm,and the single plant is extracted according to the characteristics of the three-dimensional crop itself,and the location is located,and the multi plant point cloud is segmented and identified.The coordinate transformation between the motion model of the designed agricultural small plant protection vehicle platform and the point cloud information is carried out.Finally,the driving system of agricultural small-scale plant protection vehicle can recognize crops in the field and complete the identification of passable areas.Finally,in order to verify the feasibility of this method,a series of experiments are designed according to this study.Firstly,the growth environment of corn plants in the field is simulated,and then the data are collected and processed.Finally,the positioning of corn plants and the identification of the passable area between rows are realized.The feasibility of this method in agricultural navigation is verified by the walking test in the passable area.