TY - GEN
T1 - Improved tracklet association for space debris using short-arc optical measurements
AU - Cai, Han
AU - Yang, Yang
AU - Gehly, Steve
AU - Wu, Suqin
AU - Zhang, Kefei
N1 - Publisher Copyright:
© 2018 International Astronautical Federation IAF. All rights reserved.
PY - 2017
Y1 - 2017
N2 - Initial orbit determination (IOD) for space debris is challenging, especially in the case where only optical observations, i.e. angles-only observations, are available and the optical observing arc is very short (i.e. too-short arc problem). The IOD solution to the too-short arc problem can be obtained using several short arcs of observations, or tracklets, from the same target, making it necessary to associate tracklets to targets across varying time intervals. The initial value problem (IVP) approach is a commonly used tracklet association method, but its accuracy and efficiency still has room to be improved. In this study, an improved IVP approach that determines the association by searching for the global minimum of a new loss function defined in canonical space is proposed. The use of the new loss function overcomes the problem of the multi-modal optimization in IVP for more accurate association. The proposed method was tested using real optical data collected from the Mt. Stromlo observatory. Results illustrated that the new method achieves 12% false negative and false positive rates which is better than 24% by the original IVP method. Additionally, the computational time is also reduced by the new method.
AB - Initial orbit determination (IOD) for space debris is challenging, especially in the case where only optical observations, i.e. angles-only observations, are available and the optical observing arc is very short (i.e. too-short arc problem). The IOD solution to the too-short arc problem can be obtained using several short arcs of observations, or tracklets, from the same target, making it necessary to associate tracklets to targets across varying time intervals. The initial value problem (IVP) approach is a commonly used tracklet association method, but its accuracy and efficiency still has room to be improved. In this study, an improved IVP approach that determines the association by searching for the global minimum of a new loss function defined in canonical space is proposed. The use of the new loss function overcomes the problem of the multi-modal optimization in IVP for more accurate association. The proposed method was tested using real optical data collected from the Mt. Stromlo observatory. Results illustrated that the new method achieves 12% false negative and false positive rates which is better than 24% by the original IVP method. Additionally, the computational time is also reduced by the new method.
UR - http://www.scopus.com/inward/record.url?scp=85050500682&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85050500682
SN - 9781510855373
T3 - Proceedings of the International Astronautical Congress, IAC
SP - 11526
EP - 11538
BT - 68th International Astronautical Congress, IAC 2017
PB - International Astronautical Federation, IAF
T2 - 68th International Astronautical Congress: Unlocking Imagination, Fostering Innovation and Strengthening Security, IAC 2017
Y2 - 25 September 2017 through 29 September 2017
ER -