Optimal Pose Design and Trajectory Planning for Close Proximity On-Orbit Inspection of Large Space Structures

<p dir="ltr">Recent achievements and goals in outer space exploration have emphasized the growing importance of autonomous in-space robotics. To pursue envisioned robotic missions in space, on-orbit servicing is found to be essential. Autonomous on-orbit inspection is a critical part of servicing and one of the capabilities required for space explorations. The inspection ensures safety in proximity operations, collects critical data, and prepares the servicer spacecraft to initiate other servicing activities. Optimal inspection planning is a challenging research topic and has received little attention thus far, especially for space applications. Therefore, this research study aims to particularly address a full 6-DOF motion planning problem for full-coverage inspection in close proximity of space structures by a single or collection of small spacecraft. A complete inspection is desired, that is, a close and continuous observation of every point on the space structure. To this end, a kinodynamic full-coverage sampling-based technique called Random Kinodynamic Inspection Tree, well-suited for inspection planning for cluttered, high-dimensional environments and spacecraft with differential constraints, is developed. Then, a differentiable and convex formulation for spacecraft trajectory optimization problem is developed for a small spacecraft providing optimal and complete inspection of a target regardless of its geometry and shape complexity. To achieve optimal and complete inspection for a rigid body spacecraft, a novel 6-DOF formulation and method of solution are developed, and a pseudospectral optimal control solver is implemented to provide an optimal solution minimizing a balanced inspection time and control input. Additionally, for the inspection of large complex space structures, multiple small spacecraft are found to be effective. A decentralized multi-spacecraft methodology is developed for a cooperative collection of small spacecraft to simultaneously inspect large complex space structures. Spacecraft coordination, communication, and task allocations are developed to decrease the total inspection time. A collision avoidance technique is developed that guarantees collision-free multiple spacecraft planning.</p><p dir="ltr">Therefore, the foremost contribution of this research is developing a novel optimal full pose design and trajectory generation methodology for close and complete inspection of large space structures using small rigid-body spacecraft. This research extends its contributions by developing a robust multi-section on-orbit inspection mission design, considering practical implementations, including the implementation of decentralized coordination for a cooperative and homogeneous collection of small spacecraft and the determination of the required number of spacecraft to perform a full-coverage on-orbit inspection of large complex space structures. Consequently, the result of this research facilitates the completion of complex on-orbit inspec tion missions using a single and multiple spacecraft. The developed research is applicable to other robotic inspection mission outside the space realm and is not limited to spatial application, but the focus of this dissertation is on on-orbit inspection planning in the deep space environment to provide necessary infrastructures for the pursuit of space exploration missions.</p>