Tomoaki Sakurai
(Keio University, Japan)

Recently Space Situational Awareness (SSA) is attracting attention in Japan and Western countries. In Japan, it is recognized as a measure to secure safe space exploration in Basic Policy on Space Exploration issued by Ministry of Defense in January 2009. In the US National Space Policy of the United States of America issued in June 2010 emphasizes the need for international cooperation in SSA including space debris observation and crash avoidance. However, not many studies are done on cooperation in solving technical issues of SSA and its operation systems. This presentation focuses on observation and monitoring of space debries, which composes SSA concept, and reports the analyses results based on our study of each of functional areas by the radar system and the optical telescope system. Furthermore, we argue the importance of observation method using large-scale optical telescope in the observation step. Finally, we focus on technical issues of SSA, current cooperation and operation systems and design and suggest a practical system.

Toshifumi Yanagisawa
(JAXA, Japan)

JAXA is developing analysis methods which are able to detect faint GEO objects that are not on the catalog provided by U.S. The stacking method, which uses numeous CCD frames to detect objcets under background noise level, is developed and shown to work well, so far. However, the method has the disadvantage that is time-cosuming to detect objects whose movements are unpredictable. In order to overcome this, a new algorithm which uses binarization of CCD images and calculates sum values instead of median, is developed. Moreover, the algorithm is applied to the FPGA board system. These reduce anlysis time to about one thousandth which enables us to analyse one night data till next night observation. Another fast analysis method is the line-identifing technique which will find straight-lined candidates on time-sequential CCD frames. This method can also detect faint GEO objcets. Although the stacking method can detect fainter objects, the analysis time of the line-identifing technique is much faster than the stackimg method. The technique can detect 40cm-size objects in GEO using a 35cm-telescope. These two analysis methods may contribute to detecting many un-cataloged objects in the near future.

Koki Fujita
(Kyushu University, Japan)

This study proposes a novel image tracking technique to detect faint debris images in the observatory image sequences obtained in the Geosynchronous Earth Orbit (GEO) region. In order to deal with the same purpose, a couple of effective techniques such as stacking and line-identification were proposed in the past studies. In spite of the effectiveness of these techniques with their detecting performance for the debris^' positions and motions in the image sequences, they need to be applied in a complementary style for these parameters. Furthermore, there is a practical issue for stacking, not necessarily effective with its computational load. This study derives a method to simultaneously detect the debris^' positions and motion vectors from the image sequences by using a computer vision technique called an optical flow algorithm. The new method detects these parameters in affirmative computational time. This study also discusses the image motion of the objects in the GEO region to confirm the feasibility of the derived technique. The effectiveness of the method is validated by applying the real image sequences obtained at the Mt. Nyukasa optical observation facility.

Masahiko Uetsuhara
(Kyushu University, Japan)

Breakup events of space objects give the biggest contribution to increase the debris population in near earth orbits. Ten or more explosion events have been suspected in the Geostationary Earth Orbit (GEO) region, but only two of them have been confirmed by actual observations. To reveal the debris population related to breakup events effectively, this paper introduces a search observation method, which enables us to detect and identify objects from a target breakup event by ground-based optical observation. This paper targets on fragments from the Titan IIIC Transtage (International Identification of 1968-081E) breakup event of February 21st 1992. This event has taken place near GEO to release at least 23 fragments, all of which have been missing now. The search observation utilizes population and motion predictions of debris fragments generated from the breakup event. Population prediction calculates the presence probability density of fragments in equatorial coordination systems. Motion prediction calculates the motion probability density of fragments in spatial temporal images acquired by Charge-Coupled Device (CCD) sensors. Population and motion predictions give unique aspects that assist the observation planning, object detection and correlation procedures. This paper also reports 1968-081E search observation results.

Moriba Jah
(Air Force Research Laboratory, USA)

To date, work has been performed in autonomously fusing photometry and astrometry exploiting multiple model hypothesis testing with encouraging results for simultaneous resident space object (RSO) tracking and characterization. Photometric and astrometric data are complimentary in that they are sensitive to different aspects of the problem: photometry is sensitive to RSO characteristics and fairly insensitive to RSO orbit, whereas the astrometry is the opposite. The research demonstrates the ability to exploit these data to uniquely identify and discriminate these inactive RSOs whether they are alone or in a clutter. Research has also been performed in more accurately recovering and predicting the actual RSO state and parameter errors. This is now combined with the data fusion in order to enhance and improve the tracking and characterizing of inactive RSOs. A comparison between traditional Gaussian error assumptions and the Adaptive Entropy Gaussian Information Synthesis (AEGIS) method for approximating the actual error probability density function is demonstrated.