Designing computationally light algorithms for concurrent real-time attitude estimation and sensor calibration

Date

2023-5-23

Author

Benli, Doğukan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

216
views

109
downloads

Cite This

Computer processors’ computational power and efficiency continue to improve each year. This improvement enables the usage of less power for computing or more complex models for embedded systems, especially in the aerospace industry. However, implementing newer and more advanced technologies to the systems in the field or production can be challenging. Therefore, focusing on more capable or efficient algorithms for these systems is still crucial. This research aims to develop and evaluate methods that decrease the computational load of onboard attitude estimation algorithms that use the Kalman Filter (KF) as the core algorithm. Computationally efficient algorithms are essential for small satellites as they are limited in hardware and power consumption. A real-time runnable accurate attitude estimation algorithm that also estimates the additional parameters, such as the sensor errors, can be beneficial for these resource-limited satellites. This thesis focuses on reducing the number of KF updates without compromising the performance of the KF and demonstrates its performance on the small satellite attitude estimation problem. This task requires manipulating the measurements to get a pseudo-measurement of slower frequency. To this end, “integrated measurements” are suggested to replace the original measurements.

Subject Keywords

Attitude estimation, Computational cost, Integrated measurements, Multiplicative Extended Kalman Filter, Sensor calibration

URI

https://hdl.handle.net/11511/103263

Collections

Graduate School of Natural and Applied Sciences, Thesis