Research On Object Identification And Orientation For Autonomous Truck Loading Of Robotic Excavator

With the development of science and technology, the intelligentization of vehicles is becoming the tendency of the vehicle technology development, while the development of construction machinery is getting along with the tide. As for the Excavator, roboticized hydraulic excavator which is the key technology in the national high and new-technology industries research and development plan, is listed in the program of The National High Technology Research and Development Program (863 Program). In recent years, some robots for construction have appeared in succession, but they are not commercial on account of the high cost. The robot product for construction appears nothing in our country, and the research is still in the experimental stage. The paper focused on the research on object identification and orientation technology of roboticized hydraulic excavator working process. At first, the paper expatiated the whole frame of the object identification and orientation technology, the following chapters put great importance on multi-scale edge detection arithmetic based on the wavelet transform. The basic research has been done in this paper for the development of object identification and orientation technology of roboticized hydraulic excavator working process.Object identification and orientation technology of excavator has profound significance for its self-determination work. The key problem of the technology is automatically identifying the target of operation and establishes the relative position between working bucket and the target as the important parameters of bucket control. The technology involved many research subjects, so the research content of the paper focuses on edge detection arithmetic research, which is used to feature extraction. The paper constructs a kind of effective wavelet—second order B-spline wavelet , which is much fits for feature extraction, and concludes multi-scale edge detection arithmetic based on the wavelet. The whole paper consists of six chapter, We will make idiographic introduction as following.Chapter One IntroductionThe chapter mainly treats with the phylogeny of the excavator and its trend of intelligentization development. In addition, the chapter introduces the development current state at home and abroad of roboticized hydraulic excavator; At last, the author sets forth the main research content of the paper.Chapter Two Summarization of target identification and orientation technology of roboticized hydraulic excavator The chapter mainly demonstrates the whole technical frame of target identification and orientation technology. Discoursing upon the basic conception of machine vision and theory frame, the acquire method of depth map, the methods of target identification etc.Chapter Three Arithmetic base of wavelet analysisThe chapter mainly demonstrates the basic conception of wavelet and wavelet transform, the main member of wavelet family and the arithmetic base of image wavelet transform etc. The work of this chapter lays foundation on research on image processing arithmetic based on wavelet transformerChapter Four Edge detection for the object of excavatorAfter introducing much main edge detection arithmetic, the chapter discourses upon a kind of effective edge detection arithmetic based on wavelet transform. The chapter applies the MATLAB program simulating the actual process of image processing, analysis all the edge detection arithmetics. The results indicate that the multi-scale edge detection arithmetic based on the wavelet transform is the best processing arithmetic. Chapter Five Conclusion Chapter Five is the summary of the whole dissertation, the main research achievements are showed in this chapter.