Research And Implementation Of Caterpillar Simulator Evaluation System Based On Expert System

With the development of national defense modernization,the level of mechanization and automation of the equipment in the military has also increased.Therefore,in the mechanized warfare of the military,it is very necessary for soldiers to use the equipment masterly,and thus the training of soldiers' use of equipment plays an important role.However,in the current training evaluation,the manpower assessment is the dominant force,and because of the knowledge and mood of each coach,it is inevitable that there is a deviation in the evaluation.During the training process,because of the shortage of coaches,the coaches cannot accompany the trainees for one-on-one exercises.At the same time,it is impossible to equip every trainee with an expert coach.If trainees are allowed to practice and familiarize the operation process by themselves,it will inevitably lead the consequences to be inefficient.Therefore,in this topic,an evaluation system has been designed.It gives the trainees more opportunities to make corrections.Let the assessment system accompany trainees to train and learn.In this way,trainees can better understand their strengths and weaknesses.In this article,the following work has been done:(1)According to the needs of the evaluation system,the training evaluation system has been designed and divided into four major modules: the expert and trainer parts,the driver part,the server part,and the crawler simulator communication hardware part.Then we analyze the function of each module and design its main function.(2)The composition of the expert system includes six parts: knowledge base,human-computer interaction interface,inference engine,comprehensive database,interpreter and knowledge acquisition.Secondly,comparing with the mainstream expert system,taking into account the clear rules of evaluation training tests and the relatively fixed steps,an expert system that is based on production rules should be used.(3)In the eclipse development environment,the human-computer interaction interface of the evaluation system is written by using the swing component in java,which is more in line with the daily computer operating habits of the general public.Basically,the mouse and the numeric keypad can complete the operation.(4)The human-computer interaction interface mainly completes the functions of maintaining the knowledge base,viewing personal historical scores,operational records,and the rankings of various subjects.(5)In the visual studio development environment,C# is used to call the iCAN master function library and fieldbus technology is used to collect and process the trainees' s operation steps,then C# is used to connect to the MySQL database.Each operation step of the trainees will be stored in a database for post-inference engine scoring.(6)In the design of the core knowledge base,we proceed from the production rule expressions and finally use the graph theory in the data structure to store the acquired knowledge.(7)The another core of the expert system is the inference engine.In the design of the inference engine,the reasoning direction and inference strategy are mainly taken into account.Finally,the inference process and the deep optimization search are used to complete the reasoning process.Using this assessment system,trainees can not only view the score of the operational flow,but also learn the specific error types about each missing point.Secondly,trainees can check their past historical results and operation records,then generate an assessment log to make a vertical comparison of their historical results.Trainees can also view the rankings of each subject in the whole class,so that they can make a horizontal comparison.This can help them understand their own weak points,and make the training effect more effective.