Summary: | 碩士 === 國立成功大學 === 航空太空工程學系碩博士班 === 101 === This study targets at the application of GPS in the CCD camera automatically tracking using the GPS receiver as the basis for observations. Through digital signal processing with CCD control mechanism as well as unmanned aerial vehicles, this study can further achieves development on platform of air monitoring system integration. As the current location information can be obtained by using GPS receiver on unmanned aerial vehicle through RS232 signal transmission to the signal processor DSP and the measurement equipment on the vehicle. The output for a PWM servo motor is calculated from its current positions and the location of target fixed coordinates to move the mechanism through them motor CCD camera to keep tracking of the action. Thereby it enables the CCD image transmitted to constantly target at object, so that the ground-based station can observe the target constantly.
The overall framework of the airborne system mainly consists of dsPIC single-chip design, with a GPS receiver and servo motor as well as CCD image transmission system. And flight control system is made by the ArduPilot which needs to match with IMU shield, ArduPilotMega and other kinds of peripheral sensing element. Through comprehensive flight calibration and control provided by the Mission Planner software, communications equipment is used to send data back to the ground station for the desired observation. After the flight test by a quadrocopter with this tracking system, and trackability defined as shown on the screen of the image, the overall system trackability can approach more than 95%. It’s a successful system in tracking the target in a fixed position and transmission the ground station for observer.
In this study, the CCD automatic tracking system on unmanned aerial vehicle is implemented though not perfect, and it is impacted from various conditions of flight information, such as a gust or wind shear which may influence the accelerometers and gyroscopes, etc., so that the resulting image will be a little biased. Consequently, there are still rooms for improvement in future study.
|