Autonomous determination of unknown characteristics of an asteroid using optical and altimeter information
- Paper number
IAC-06-B3.P.1.01
- Author
Mr. Yufei Liu, Harbin Institute of Technology, China
- Year
2006
- Abstract
In recent years, the deep space exploring has been attached to great significance. As to the deep space exploring mission, such as flyby, rendezvous, landing and returning samples to an asteroid, it is important to get detailed information in regard to the shape and the motion characteristics of the asteroid in advance. During the approaching process, the distance information can be obtained by the altimeter in that the distance is quite close. At the same time we can also obtain images of the target body using the onboard navigation camera. In the successful STARDUST mission that is to flyby the comet and return samples, the navigation algorithm uses the image of target body to obtain nucleus center of figure data and then update the spacecraft position and attitude. Based on the successful algorithm, a simple, reliable, and fast algorithm is developed to estimate the direction of spin axis, the spin rate and the feature point radius. Here we assume that during the approaching process the relative motion of the spacecraft approximates to a uniform linear motion in the inertia space, if we limit the time of observation to less than 1200 seconds. Therefore the motion of the spacecraft can be considered to be uniform linear motion to the asteroid. Based on the assumption, it is valid that after observing the feature point four times at the same interval, we can get four observation vectors which can express the relative position and velocity of the spacecraft and the spin rate and feature point radius of the asteroid. So it is feasible that the information obtained from images of the feature point and the altimeter in the course of approaching can be processed by the algorithm, through two different filters to reconstructing the target relative state information and get the characteristics of an asteroid: a batch least squares filter and an extended Kalman filter. The strategy and the effectiveness of the two filters are compared. One method to check the observability of the system is to use a numerical method (check the observability matrix conditioning) performed by Jo Ryeong Yim. The other is to use the geometrical relationships between the observation vectors and the states. The validity of the proposed method is verified by performing the Monte Carlo simulations. These simulations incorporate errors in the initial states. The results prove that the algorithm is successful in estimating the spin rate and radius of feature point. The results also turn out that the estimation error is less than ten percent. We also find that the algorithm is sensitive to the altimeter information. First, this paper introduces the sensors for the algorithm, which are the camera in STARDUST mission and the laser altimeter in NEAR mission. Second, a summary of the state and observation equations are given. The state includes relative position and velocity in inertial coordinate, spin rate and feature point radius. The observation equations use four vectors from the camera to the feature point. Third, the extended Kalman filter and a batch least squares filter are reviewed. Fourth, an analysis of the observability of the system follows. Finally, Monte Carlo simulation results and discussions are given.
- Abstract document
- Manuscript document
IAC-06-B3.P.1.01.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.