Research On Program Generation Based On AORBCO Model

Due to the increase in computing speed and storage capacity of computers,the amount of data processed by them is increasing,the data structure is becoming more and more complex,and the algorithms for processing data are becoming more and more complex,making the application of computers more and more extensive,and it seems that computers are becoming more and more intelligent,but the basic principles of computer program storage and program control have not changed.From the analysis of the research field of software intelligence,the intelligence displayed by computers is reflected in the automatic execution of programs;human intelligence is reflected in the design and development of software;artificial intelligence should mainly be reflected in the automatic generation of programs.The research purpose of automatic program generation is to enable computer software to have a certain perception and understanding of the environment in which it is located,and to automatically generate solutions to achieve the established goals according to its current state and ability,and execute the solutions to achieve the goals,and Learn more knowledge and abilities in the process.First of all,this paper studies the human intelligent structure and the related intelligent mechanism of humans to solve the problem,and analyzes the composition of the program and the human programming process according to the intelligent structure and intelligent mechanism.Through the analysis and research of the AORBCO model and its description language,the AORBCO description language is used to describe the knowledge of Ego,which adds semantics to the data and facilitates the computer's understanding and calculation of the data.To enable Ego to automatically generate programs,it is necessary for Ego to have a certain understanding of its own capabilities,and use the axiomatic semantics in formal semantics to describe the semantics of the procedural knowledge in the model,that is,through the preconditions of actions,the description of post-conditions defines the semantics of the action.Secondly,the program generation in the AORBCO model is a process in which Ego uses its own beliefs and abilities to achieve a desire,that is,the Ego planning process.Intelligent planning is an important part of the field of artificial intelligence,the research and development of the planning process mainly focus on heuristic search methods and planning model design methods.How to design a general heuristic function is still a big challenge.Summarizes the process of describing,decomposing,and synthesizing problems in human program design,semantic description,hierarchical division,and organization of computer capabilities,and proposes update and evolution theories.In addition,the Java programming language was used to analyze and design the planning problem,and the design path planning experiment was carried out to verify the components of the plan.Analogous to the scenario where humans solve problems,when they cannot solve the problem due to their lack of ability,the self-aware Ego can solve the problem by communicating with acquaintances in the belief.When Ego conflicts with acquaintances or acquaintances,the problem can be resolved through negotiation.When Ego communicates with the acquaintance in the belief to solve the problem,the two parties negotiate the format of the requested data,the method of the request,and the form of the planning result to deal with the problem.Aiming at the negotiation conflicts and deceptive behaviors in traditional intelligent planning problems,this paper proposes Quality of Planning(QoP)verification to solve them,and updates and evolves the negotiation results and weights.Finally,use Java GUI,JADE,graph database related technologies to design and implement the components of the intelligent system,and design a planning system to analyze and verify the problems.By systematically solving the graphic area problem,the intelligent process of human solving the problem is simulated,and the ability to solve a specific problem is verified,and the rationality and feasibility of the system's automatic planning are verified.