A new parameter identification strategy of parachute inflation phase based on monocular video

A significant number of videos and images are recorded during parachute drop tests. However, due to the limitations of the test conditions, the majority of videos are monocular and of relatively low quality. Also, these videos are mostly used only for image records and qualitative analysis, lacking application in parametric quantitative analysis of the parachute. Parameters of parachutes, such as the changing of area and motion characteristics during parachute inflation and deployment, can be included in the video analysis. These parameters play vital roles in dynamic modeling, simulation analysis, and performance evaluation of the parachute system. This paper proposes a parachute parameter identification strategy based on the monocular video and the characteristics of the parachute. This strategy includes four steps: coarse positioning of the canopy, fine segmentation of the canopy, solution of scaling factors, and solution of parachute parameters, with the foundation of camera imaging principles and image processing methods. By analyzing and verifying the video images from an airdrop test of the parachute for lunar automatic sample return, the results show that the tilt angle of the canopy, canopy perimeter, and parachute projected area can be identified using this method, with the identification errors ranging from -2.632 % to 4.031 %. Therefore, the proposed parachute parameter identification strategy is reasonable, feasible, and accurate. Relevant results will also significantly impact the theoretical research and practical engineering design of parachutes.