Research And Application Of Compiling Methods Based On Graphical Programming

Robot programming is mainly for non-computer professional users,but programming based on text is difficult to meet user programming requirements.The graphical programming language is generally used for robot programming,and then the final executable file is generated by the special compiling system.At present,the cross-compiling method is usually used to complete the compilation of the graphical program,which includes a pseudo-code generation process and a cross-compiling process.However,the existing iterative analysis algorithm is difficult to satisfy the syntactic analysis of this paper.At the same time,the cross-compiling method is inefficient and target machine code generated by this method that has a complex structure and a large space,so it can not meet the compiling requirements of educational robot.Aiming at the above problems,a simple,efficient and retargetable direct-compiling method is proposed in this paper.It is based on the syntax and semantics of the graphical programming language to directly convert the graphical program into the target machine code.Syntactic analysis is a key problem.This paper focuses on the syntax and the compiling methods of graphical programming languages.The main tasks are as follows:(1)Problem description and mathematical modeling: The compiling problem in the graphical program is discussed,and the graphical module and the program structure are abstracted and modeled according to the direct-compiling requirements to fully describe the data constraints and parallel structuresin in the graphical program,and discuss the related definitions and characteristics of these two models in detail.(2)Iterative analysis Algorithm analysis and improvement: Analysis of data constraints and parallel structures in the graphical program,aiming at the deficiencies of existing iterative analysis algorithms,this paper proposes an iterative analysis algorithm based on constraint separation,which includes thread merging algorithm and signal generation algorithm.The thread merging algorithm determines the thread information,the semaphore generation algorithm determines the synchronization information between the threads,and then combines the two section in the code generation phase to satisfy the data constraint and module parallelism.(3)System implementation and experimental analysis: The DAG diagram model and the iterative analysis algorithm based on constraint separation are applied to the compiling system of this paper.Its design framework and implementation process are discussed in detail.The code generation effect and robot movement effect are analyzed to verify the correctness of the algorithm.At the same time,statistical analysis is performed on compiing time and code space to verify the efficiency and the optimization ability of the code of the direct-compiling method.