GPS Remote Sensing Image Technology Pepper Picking Robot Intelligent Navigation And Trajectory Planning Research

With the advancement of smart agriculture and the rapid development of mobile robots,how to use robots to solve the difficulties encountered in the development of current agricultural production is the focus of current agricultural robot research.The prickly ash picking robot studied in this paper is based on the problems of low picking efficiency and difficult picking in the current manual picking.Research on the autonomous navigation and trajectory planning of the Zanthoxylum robot.Starting from the identification of the prickly ash forest,obtain the environmental information of the prickly ash tree,determine the position of the prickly ash tree trunk,and comprehensively consider the length and safety of the path,plan the target path of the prickly pear picking robot mobile car,build an experimental platform,simulate the picking environment,and realize the Autonomous navigation and trajectory control of pepper picking robot.(1)Information extraction of prickly ash forest based on GPS remote sensing images.Compared with the traditional forest land information extraction method,GPS can be used to locate the object space of Zanthoxylum bungeanum and determine the range boundary.Combined with high-resolution remote sensing image data,the extraction of Zanthoxylum bungeanum information is fast,convenient and efficient.After the rough picking range was determined by GPS,the image-based object-oriented classification method in ENVI was used to extract the information of the prickly ash forest,and the prickly ash tree crown was extracted;the tree crown information was vectorized and imported into ArcGIS and the elements were The centroid point of the Zanthoxylum bungeanum tree trunk was obtained by quasi-exchange,and the extracted centroid position was combined with the WGS-84 coordinate system to obtain the Zanthoxylum bungeanum tree trunk coordinates and combined with the GPS map,which provided a path planning target for the mobile car.(2)Picking path planning of the mobile car of the pepper picking robot in the static environment.Aiming at the static environment of many prickly ash trees in the prickly ash forest,an improved ant colony algorithm is proposed for the path planning of the prickly pear picking robot mobile car in the global static environment.Ant colony algorithm has the defect of blind search at the initial moment,and a non-uniform distribution model of pheromone is established for this;for the selection of path nodes,a certain probability of randomness is increased,the probability of certainty is reduced,and the diversity of search is increased;Aiming at the node selection problem of ant colony algorithm in the moving process,taking the current local node as the reference strategy,this paper introduces the relationship between the current node,the next node and the end point,and the search path length is saved from a global perspective,and the planning efficiency is improved.By introducing the force in the potential field method into the heuristic factor,the safety of the planned path is improved to avoid collision with obstacles;considering the stability and safety of the car when driving,the path smoothing and redundant point removal strategies are introduced.Redundant point removal not only shortens the path length but also reduces unnecessary inflection points,and the smoothing of the path increases the stability of driving and improves stability.(3)Picking path planning of the mobile car of the pepper picking robot in the dynamic environment.Aiming at the uncertainty in the actual environment of the pepper forest,especially the collision of the car caused by the appearance of the unknown obstacle and the dynamic obstacle,the dynamic path planning of pepper picking robot mobile car based on dynamic window method and improved ant colony algorithm is proposed.According to the integrated algorithm proposed in this paper,path planning,firstly,the mobile trolley of the pepper-picking robot tracks the path planned in the global static environment.During the forward movement,the sensors carried by the mobile trolley continuously update the local environment in real time.When there is an obstacle,the mobile car avoids collision with the obstacle by performing local path planning.After completing the local path planning,adjust the car’s posture and continue to drive according to the original path.In this way,through the combination of the global path and the local path,the realization of It realizes the path planning of the trolley in a dynamic environment and safely reaches the target picking position.(4)Carry out the experimental research on the autonomous navigation and trajectory planning of the prickly ash picking robot equipped with the experimental platform.Design and build a prickly pear picking robot mobile trolley,select control hardware,set the experimental environment,place the trolley in the experimental environment,and simulate the path planning experiments of the trolley in static and dynamic environments by positioning the prickly pear tree picking target,respectively.Feasibility of trajectory planning.