Font Size: a A A

Research On Key Technologies Of Scene Simulation And Visual Navigation For Space Target Photoelectric Detection

Posted on:2021-01-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:K YuFull Text:PDF
GTID:1488306569983699Subject:Optical Engineering
Abstract/Summary:PDF Full Text Request
Photoelectric detection technology is a commonly used information acquisition method in space target tasks and an important development direction in the field of space situational awareness.The development of a new generation of space target photoelectric detection system has the characteristics of multi-spectral band,multimode,integration,etc.In the design,testing,evaluation and other stages of system development,digital simulation technology is needed to provide tool support.Visual navigation is an important application direction of photoelectric detection technology.Space service tasks need the support of visual navigation methods for space targets with strong adaptability,high measurement accuracy and fast processing speed.Based on the digital technology requirements in the development of China's space target photoelectric detection system,photoelectric detection scene simulation technology and space target visual navigation technology were investigated deeply and systematically in this dissertation.The research results in scene simulation have been applied to the testing and evaluation of some space target photoelectric detection systems.The research results in visual navigation provided a ground simulation measurement technology support for the development of China's space service navigation technology.The main contents of this dissertation are organized as following:Research on simulation method of space target imaging.The ray tracing method was used to realize the simulation calculation of the space target physical imaging.The radiation characteristics of skin and exhaust plume were modeled.A method for optimizing the calculation of the reflected radiance of the skin with direct lighting multi-importance sampling was proposed.A method for optimizing the calculation of the outgoing radiance of the exhaust plume with combining integration was proposed.The infrared radiation characteristics of stealth aircraft skin and exhaust plume in different detection scenarios and different spectrum were studied.The main influencing factors and suppression laws of infrared radiation suppression of solid particles on exhaust plume were simulated and analyzed.Research on simulation method of detection background imaging.From the measured images,the horizontal and vertical distribution of the radiance on the limb background was studied.A statistical model of the infrared radiation characteristics on the limb background was established.A fast simulation method of infrared images on the limb background was proposed,and the accuracy and effectiveness of the method were verified through simulation experiments.Models for imaging positioning and radiation characteristics of stars were established.Based on the star catalog database,the fast simulation method of the deep space star background images was realized.Research on overall design and integration of photoelectric detection scene simulation system.Through the research and analysis of the photoelectric detection scene simulation software abroad,the overall structure,main functions and key technologies of the system were designed.The general scheme of data model class and simulation calculation flow of simulation system was designed.A space-based infrared detection scene simulation system for space targets and a digital imaging simulation source for space target navigation cameras were developed,and the development plan and application results were introduced.Research on visual navigation algorithm of space cooperation target.Based on the research results of the photoelectric detection scene simulation technology,the digital imaging simulation source software system for space target navigation cameras was used in the research work of the visual navigation algorithm of space cooperation target.By establishing a cooperative projection relationship between the3 D model and the measured images on the edge contour,the application scope of the cooperative target visual navigation technology was expanded.Using the "offline template generation + online search matching" scheme,the target pose calculation based on the search matching method was realized.The 2D image search matching process and the 3D pose search matching process were accelerated through the image pyramid.The accuracy,stability and calculation efficiency of the method were verified through simulation experiments.Research on visual navigation algorithm of space non-cooperation target.Based on the research results of the photoelectric detection scene simulation technology,the digital imaging simulation source software system for space target navigation cameras was used in the research work of the visual navigation algorithm of space non-cooperation target.The docking ring was used as a reference feature for target recognition and pose calculation.A method for contour extraction of docking ring was proposed for the measurement image in complex noise environment.The binocular vision measurement system was used to achieve the target 6-DOF pose calculation based on the contour of the docking ring and the reference feature point outside the ring.Based on the geometric symmetry and the shortest path principle,the capture point on the docking ring was determined.The capture position and capture angle were calculated by a monocular camera and a laser rangefinder.The accuracy and stability of the method were verified through simulation experiments.
Keywords/Search Tags:space target photoelectric detection, imaging simulation, visual navigation, ray tracing, simulation system design, contour feature
PDF Full Text Request
Related items