In recent years，space situation awareness has garnered increasing attention in both military and civilian fields.Space-borne inverse synthetic aperture radar （SBISAR） is an important tool for space situational awareness.It enables high-resolution imaging of space targets，and supports subsequent tasks such as component extraction，posture estimation，and type identification.However，in SBISAR imaging scenarios，complex and non-uniform high-speed relative motion exists between the platform and the space maneuvering targets，rendering traditional range-Doppler （RD） imaging methods ineffective.Therefore，motion compensation should be considered. In this paper，two typical motion errors are discussed，i.e.，the failure of the traditional “stop-go-stop” model under high radial velocities and the inconsistent target rotation speeds within long coherent processing intervals （CPIs），and a specific algorithm is designed for intra-pulse and inter-pulse motion compensation.First，a signal model for SBISAR imaging is established.Then，based on the minimum entropy criterion，a parameter estimation cost function is constructed to estimate the chirp rate of the echo signal，and the obtained estimation is used to construct a compensation term for intra-pulse motion compensation.Similarly，an image entropy cost function is established to estimate the rotational parameters，achieving spatial-variant phase compensation for rotational motion.The proposed algorithm can effectively address the range extension and azimuth defocusing issues by traditional RD imaging algorithms in the case of high-speed maneuvering，and enables the high-resolution imaging of space targets by SBISAR.Through the experimental analysis based on the simulation data，the effectiveness of the proposed algorithm is validated.