Research Based On Dual Quaternion For Relative Position And Attitude Estimation Of Formation Flight Satellite

The technology of formation flying is a hot research topic in the field of aerospace currently, as itis a novel approach for space mission to enhance the overall performance compared with theconventional single satellite approach. The premise of formation flying is precisely known relativeposition and attitude. Pose estimation is the core problem would be solved for satellite formationflying, rendezvous and docking, autonomous navigation space science and technology and many otherengineering practice.The general spatial rigid body motion is completely equivalent to coordinate system’stransformation. Comparing with traditional quaternion description, dual quaternion can represent therotation and translation in a unified pattern, and more concise and easier to operate. Compare to otheroptimization, if convex optimization model there is a local minimum, the minimum point is the globalminimum point, therefore, the use of convex optimization modeling can avoid algorithm fall into localminimum.Inspired by this fact, this paper presents a novel vision-based relative position and attitudeestimation algorithm using points to regions correspondence for close formation flying. Given thepoints on the leader satellite surface and the convex regions in which the correspondent image pointslie on the follower satellite image plane, combining dual quaternion and convex optimizationmathematical tools to take full advantage of the simplicity of the dual quaternion description of thecoordinate system transformation, the convex optimization problem representation is formulated byusing dual quaternion representing the rotation and translation, by means of which the relativeposition and attitude parameters are estimated. This algorithm uses the advantage of the dualquaternion description which is better than the traditional quaternion coordinate systemtransformation, and using convex optimization can reduce the requirement of precise point to pointcorrespondence in the traditional algorithm. Simulation and experiment shows that the algorithm canmeet the demand of relative pose estimation accuracy and has a good robustness for close formationflying, and this algorithm can provide a reference for vision relative pose estimation between variousspacecraft in the current space mission.