研究目的
To develop and implement a reliable star identification algorithm for low-cost star trackers, addressing sensitivity to missing stars, efficiency to magnitude noise, and identification failures with few stars detected.
研究成果
The proposed algorithm improves identification rates, robustness to missing stars, and efficiency to magnitude noise with a reduced database size. It is suitable for low-cost star trackers and shows promising performance when implemented on ARM hardware, but requires additional validation steps for false stars.
研究不足
The algorithm is sensitive to false stars and requires a verification step for robustness. It relies on simulations and may need further validation with real-world testing. Hardware implementation is on a prototype, not a final product.
1：Experimental Design and Method Selection:
The study uses a modified Liebe algorithm with new strategies for star triplet selection and reference star criteria. A sky simulation program is developed to test robustness under noise conditions.
2：Sample Selection and Data Sources:
Simulations use the Hipparcos Star Catalogue and generate random camera orientations.
3：List of Experimental Equipment and Materials:
Includes a prototype Data Processing Unit (DPU) with ARM Cortex-M4 processor, external memories (Flash, SRAM, FRAM), and simulation on a PC (ThinkPad X240).
4：0). Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: Simulations involve adding Gaussian noise to star positions and magnitudes, inserting false stars, and removing stars. The algorithm is implemented on hardware for performance testing.
5：Data Analysis Methods:
Identification rates are calculated, and performance metrics (memory usage, running time) are analyzed using C language and simulation software.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
