Answers are grounded in this paper and cite their sources.
This paper considers the problem of determining agent positions from which to perform a multi-agent inspection of an unknown RSO (resident space object) that is freely tumbling. Uncertainty in rigid body dynamics combined with multiagent coordination make this problem difficult to optimally solve. We propose a Monte Carlo methodology to assess the performance of simple passive strategies. This can be used as a benchmark that more complex strategies should aim to exceed. Two different passive agent placement strategies are considered: 1) constant relative coordinates with respect to the RSO or 2) stable periodic trajectories requiring no actuation. The tumbling satellite dynamics are approximated using Euler's torque-free rotation equations where the inertia matrix is unknown and the angular velocity and attitude are uncertain. We assume the unknown rigid body has a set of equispaced keypoints we would like to inspect. The inspection problem is to then find the best locations to place agents such that the most keypoints are inspected on average despite uncertain dynamics. First we present a Monte Carlo based approach for determining keypoint coverage estimates for multi-agent passive inspection. Second, to support our Monte Carlo methodology, we develop a scheme that both filters and estimates parameters needed to predict rigid body behavior.
No takes yet. Share an insight, caveat, or question.
Cutlip et al. (2024) studied this question.