Identifying pitching mode of aerial planting projectile by the use of image processing. (December 2017)

• Record Type: Journal Article
• Authors: Goodarzi, H; Sabzehparvar, M
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 231:Number 14(2017)
• Page Start: 2616
• Page End: 2633
• Publication Date: 2017-12
• Subjects: Aerial plantation -- system identification -- image processing -- dynamic simulation; Aeronautics -- Periodicals; Astronautics -- Periodicals; Airplanes -- Design and construction -- Periodicals; Aerospace industries -- Periodicals; 629.1
• DOI: 10.1177/0954410016668908
• Languages: English
• ISSNs: 0954-4100

Abstract

Aerial planting is a new and affordable method, which is used for purposes of reforestation and restoring pastures. In this study, flight dynamics of an aerial planting projectile has been simulated using data of the wind tunnel test. By creating the projectile graphical model and using the results of dynamic equations for this model, pitching angle has been simulated graphically by recording the pitch attitude of the graphical model and, by applying spectral filtration the rate of pitch angle at any point of the trajectory is calculated. Primary dynamic model parameters including the coefficients of regression equation of motion are estimated by applying the least squares estimator. Comparison of the estimated coefficients from image processing and coefficients given to the simulation program reveals high accuracy of the model resulting from the image processing data used in this method and approves wind tunnel test data by graphical model and calibration of camera measurement. Also, an experimental test setup has been created and the images of projectile falling down in the presence of fan flow have been captured with high-speed digital camera. By decreasing the ambient light intensity, the images and measurement noise of theta angle increases. By applying recursive least square algorithm, sensitivity of coefficients and robustness of this algorithm has been analyzed. The analysis results indicate that the estimation of coefficients using image processing data has great accuracy.