Font Size: a A A

Based On Webots Crane Work Scene Perception And Path Planning Research

Posted on:2018-03-26Degree:MasterType:Thesis
Country:ChinaCandidate:L Q LiFull Text:PDF
GTID:2322330518479141Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
The lifting process of the crane is complicated and complicated,and the lifting scheme is very difficult.The safety of the hoisting scheme directly affects the safety and cost of the hoisting,and the automobile crane takes advantage of its own,occupying a great proportion in the hoisting work.Faced with increasingly complex lifting and lifting environment,relying solely on manual experience to complete the lifting work,so that the safety and efficiency of lifting will be reduced.In order to solve the problem of hoisting and improving the hoisting safety,the simulation technology based on virtual hoisting came into being.Based on the Webots simulation software,this paper studies the work scene perception and path planning algorithm in the three-dimensional virtual hoisting environment of the crane,and finds out the relationship between the lifting objects,the obstacles and the cranes.An optimal path to ensure safe lifting.The main tasks include:Firstly,the actual hoisting technology,main technical parameters,basic operation,sensing system and hoisting path parameters of automobile crane are analyzed in detail.Through the analysis of the factors affecting the lifting path,the model of lifting environment is established in Webots.Through the crane's perception system,Get the scene of lifting the scene,the lifting of the scene to carry out information on the scene,for the path planning to prepare.Secondly,according to the crane lifting characteristics and stations to determine the lifting point and the starting point.This paper proposes a path planning algorithm based on improved genetic algorithm.By using the job scene model and the truck crane pose space model,the hoisting path space is divided into multiple path planes for processing,the path plane model is established,and finally the path information is established by the genetic algorithm The model determines the hoisting path of the crane.The algorithm improves the search precision by adding the bounding box attribute(ie,the boundary object attribute)to the scene model,and improves the computation rate and convergence speed with the memory operator.Finally,in the virtual lifting software platform,using VRML97 programming language,complete the virtual hoisting simulation test.
Keywords/Search Tags:scene perception, Webots, pose space model, improved genetic algorithm, memory operator
PDF Full Text Request
Related items