Autonomous and optimal guidance computation for mapping of spatial targets remains a major challenge due to the nonlinear nature of spacecraft dynamics, the complexity of mapping objectives, and the computational constraints of space-flight hardware. Most of the proposed strategies rely on heuristic path planning or reinforcement learning, lacking optimality guarantees and systematic constraint satisfaction. This work introduces a convex-optimization-based framework for trajectory design that reformulates the mapping task as a tractable optimization problem. By leveraging Successive Convex Programming (SCP) and a novel cost function formulation, the proposed approach enables systematic constraint handling, robust convergence, and onboard implementability, to obtain fuel-optimal solutions that comply with usual platform-related limitations of small spacecraft. The method is numerically validated in two relevant operational scenarios: the inspection of artificial objects in Low Earth Orbit (LEO) and recurrent feature mapping of asteroid surfaces, demonstrating its potential for reliable and resource-efficient trajectory planning.
Belloni et al. (Thu,) studied this question.