1: \begin{abstract}

2: The interaction topology is critical for efficient cooperation of mobile robotic networks (MRNs). 

3: We focus on the local topology inference problem of MRNs under formation control, 

4: \textcolor{black}{where an inference robot with limited observation range can manoeuvre among the formation robots. }

5: This problem faces new challenges brought by the highly coupled influence of unobservable formation robots, inaccessible formation inputs, and unknown interaction range. 

6: \textcolor{black}{The novel idea here is to advocate a range-shrink strategy to perfectly avoid the influence of unobservable robots while filtering the input. 

7: To that end, we develop consecutive algorithms to determine a feasible constant robot subset from the changing robot set within the observation range, and estimate the formation input and the interaction range. 

8: Then, an ordinary least squares based local topology estimator is designed with the previously inferred information.} 

9: Resorting to the concentration measure, we prove the convergence rate and accuracy of the proposed estimator, taking the estimation errors of previous steps into account. 

10: \textcolor{black}{Extensions on nonidentical observation slots and more complicated scenarios are also analyzed. 

11: Comprehensive simulation tests and method comparisons corroborate the theoretical findings.}

12: \end{abstract}

13: