碩士 / 國立成功大學 / 機械工程學系碩博士班 / 97 / In guidance, navigation, and control applications, precisely determination of velocity and position are essential for proper system operations. Those information can be obtained by integrating measured acceleration signals. Traditionally, using high grade inertial sensors to achieve the accuracy is the exclusive method but the cost may not be affordable for non-military or non-research applications. In this dissertation, the author presents an alternative way by using low-cost, small-sized solid state inertia sensors to replace those high-priced sensors. These low cost inertia sensors suffer from severe signal drafting after integration, which must be suppressed in order to maintain the functionality. In this thesis, the compositions of the inertia sensor outputs are analyzed first. Then, several anti-drafting mechanisms are proposed and examined; including anti-drifting integrator, heuristic signal identification filters, Kalman filters, and state update methods. The capabilities and limitations on suppressing signal drifting of the proposed methods are then evaluated by two test beds: the fixed-fixed beam system is used to examine the possibility of velocity reconstruction using accelerometers, while the linear motor dring stage is used for exploring indoor inertia navigation. The results indicate that the proposed schemes can successfully suppress the signal drifting. In particular, the scheme containing Kalman filter could achieve the best results. Finally, the proposed methods are demonstrated by using a small mobile omni-robots. The experiment result shows the mean error is 0.03m in 3.5m travel distance and 47 seconds duration and the results indicate that the proposed method could be useful in indoor positioning for smart living applications.
| Identifer | oai:union.ndltd.org:TW/097NCKU5490143 |
| Date | January 2009 |
| Creators | Wei-Cheng Lin, 林韋澄 |
| Contributors | Kuo-Shen Chen, 陳國聲 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 145 |
Page generated in 0.0124 seconds