Font Size: a A A

Rapid Control Prototyping Platform Design And Its Application On Multi-sensor Data Fusion

Posted on:2012-11-24Degree:MasterType:Thesis
Country:ChinaCandidate:B FanFull Text:PDF
GTID:2212330368482634Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
RCP (Rapid control prototyping) is a new kind of engineering desigh method that is different from traditional way, aimed to handle the problems on aircraft, spacecraft, automobile manufacture, etc. This paper designs an embedded platform based on RCP method which can be applied on MSDF (Multi-sensor Data Fusion) domain against a background of mobile robot's automatic obstacle avoidance.Analyze the advantages of RCP method comparing to traditional way and the main functions and characters of automatic code generation based on MATLAB's RTW components. Develop the embedded platform that can achieve the RCP method;Design the software and hardware of mobile robot platform; Design the communication interface between mobile robot and RCP platform and establish the communication protocol.Partition the obstacle avoidance zone and build the motion model of the mobile robot. Design the automatic obstacle avoidance algorithm with two levels structure. In the first level, design the Kalman filter with measurement fusion and rule-based expert system to pick up the characters of the obstacle. In the second level, based on rule-based expert system achive the decision level fusion and output the final control instrutions.Design the mobile robot's automatic obstacle avoidance model on MATLAB's Simulink and Stateflow environments, study the efficiency of the Kalman filter with measurement fusion and simulate the certain algorithms. Simulate the Kalman filter with measurement fusion, and test the affection of parameter R in algorithm. Finally, simulate the entire model.After the algorithm simulation, generate the embedded C code using the MATLAB's RTW function and make the praticaltality test on mobile robot's automatic obstacle avoidance function. The tests includes:Test the sensors on mobile robot and calculate the certain parameters of Kalman filter according to the test results;Carry out the mobile robot's automatic obstacle avoidance test and give out the curves on different zone during this test; Compare the results after fusion and before fusion. Finally, analyze the structure of automated generation code and evaluate the efficiency of RCP platform.
Keywords/Search Tags:Rapid control prototyping, Multi-sensor data fusion, Expert system, mobile robot, DSP
PDF Full Text Request
Related items