| Summary: | 碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 93 === The subject to be studied in this thesis is position estimation of an air-duct cleaning robot. Since an air-duct blocks radio and magnetic signals, the sensors installed on the robot are an encoder and a gyro only. The former is used to measure the speed of the robot while the latter is used to estimate its orientation. To increase the accuracy of the gyro, a hardware calibration procedure that using high-precision stepping motors will be performed first. Then different integration techniques such as the the Euler’s formula , the trapezoid rule, and the Simpson’s rule will be used to estimate the position and the orientation of the robot. Corresponding to different integration techniques, different experiments were performed and compared. In this manner, a map to describe the inner structure of the air-duct can be generated and the control of the cleaning robot can then performed according to this map. Experimental results will verify the feasibility of the proposed estimation method.
|
|---|
