Unscented Kalman Filter adaptive noise covariance selection for satellite formation flying with Q_leaming

Nemati, Mehdi; Kankashvar, Mohammad Rasoul; Bolandi, Hossein

doi:10.1109/icee55646.2022.9827301

Cited by 1 publication

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, they also adopted the astronomical integrated navigation based on the ultraviolet earth sensor and optical interferometer, and introduced QLEKF to integrate the two measurement information to estimate the position, velocity, and attitude state error of the spacecraft 36 , and assess the optical path delay bias in the optical interferometer, and the proposed QLEKF demonstrates superior estimation performance compared to the conventional EKF. Nemati et al 37 adopted the RL to enhance the position estimation of the spacecraft and proposed the state estimation algorithm based on the combination of the RL and the Kalman filter to acquire the optimal solution of the state and observation noise covariance, which improved the PVSE accuracy for the spacecraft navigation. Xiong et al 38 Based on the above researches of others mentioned, this paper presents a combined navigation scheme based on the X-ray pulsar and two-dimensional Doppler velocitymeasuring to improve the high accuracy of autonomous integrated navigation for the SSBE cruise phase, which can make up for the accumulated position estimation error in the Doppler velocity-measuring.…”

Section: / 38mentioning

confidence: 99%

“…Compared with the above research [35][36][37][38] , the intelligent Q-learning algorithm used to tune the parameters of the state and observation noise covariance is also improved accordingly. The reward mechanism and Q-table of the reinforcement Q-learning is designed according to the sub-filter characteristics of the federated filter, and the iterative period of the Q-learning is taken as the algorithm to evaluate the cumulative reward.…”

Section: / 38mentioning

confidence: 99%

“…One of the RL methods that is most frequently applied is Q-learning [46][47][48] . With the development of intelligent autonomous navigation research, Q-learning has been applied more and more in navigation system design 35,37 . Since the Q-learning is essential for adaptive trial-and-error optimization of state and observation noise error covariance matrix, we briefly describe the fusion of enhanced Q-learning and navigation information.…”

Section: Navigation Information Fusion With the Q-learningmentioning

confidence: 99%

See 2 more Smart Citations

Intelligent QLFEKF integrated navigation based on the X-ray pulsar / solar and target planetary Doppler for the SSBE cruise phase

Tao,

Song,

Lin

et al. 2024

Preprint

View full text Add to dashboard Cite

The Solar System Boundary Exploration (SSBE) is the focal point for future far-reaching space exploration. Due to the SSBE having many scientific difficulties that need to be studied, such as super-long space exploratorydistance, super-long flighttime in orbit, and super-large communication data delay between the ground and the probe, the probe must have sufficient intelligence to realize intelligent autonomous navigation. The traditional navigation schemes have been unable to provide high accuracy autonomous intelligent navigation independent on the ground for the probe. Therefore, high accuracyintelligent astronomical integrated navigation would provide new methods and technologies for the navigation of the SSBE. The probe of the SSBE is disturbed by multiple sources of solar light pressure and the complex unknown environment during the long cruise operation in orbit. In order to ensure the position stateand velocity state error estimation high accuracy for the probe cruise phase, an autonomous intelligent integrated navigation scheme based on the X-ray pulsar / solar and target planetary Doppler velocity measurement is proposed. The reinforcement Q-learning method is introduced, and the reward mechanism is designed for trial-and-error tuning of stateand observationnoise error covariance parameters. The federated extended Kalman filter (FEKF) based on Q-learning (QLFEKF) navigation algorithm is proposed to achieve high accuracystate estimation of the autonomous intelligence for the SSBE cruise phase. The main advantage of the QLFEKF is that the Q-learning combined with the conventional federated filtering method could optimize the state parameters in real-time, and obtain high position and velocity state estimation (PVSE)accuracy. Compared with the conventional FEKF integrated navigation algorithm, the PVSEnavigation accuracy of the federated filter integrated based the Q-learning navigation algorithm are improved by 55.84% and 37.04% respectively, which present with the higher accuracy and greater capability of the raised autonomous intelligent integrated navigation algorithm. The simulation results show that the intelligent integrated navigation algorithm based on QLFEKF has higher accuracy navigation,and able to satisfy the demandsof autonomous high accuracy for the SSBE cruise phase.

show abstract

Section: / 38mentioning

confidence: 99%