1: \begin{abstract}

2: Probabilistic swarm guidance involves designing a Markov chain so

3: that each autonomous agent or robot determines its own trajectory

4: in a statistically independent manner. The swarm converges to the

5: desired formation and the agents repair the formation even if it is

6: externally damaged. In this paper, we present an inhomogeneous Markov

7: chain approach to probabilistic swarm guidance algorithms for minimizing

8: the number of transitions required for achieving the desired formation

9: and then maintaining it. With the help of communication with neighboring

10: agents, each agent estimates the current swarm distribution and computes

11: the tuning parameter which is the Hellinger distance between the current

12: swarm distribution and the desired formation. We design a family of

13: Markov transition matrices for a desired stationary distribution,

14: where the tuning parameter dictates the number of transitions. We

15: discuss methods for handling motion constraints and prove the convergence

16: and the stability guarantees of the proposed algorithms. Finally,

17: we apply these proposed algorithms for guidance and motion planning

18: of swarms of spacecraft in Earth orbit. 

19: \end{abstract}

20: