1: \begin{abstract} 

2: This document contains improved and updated proofs of convergence for the sampling method presented in our paper ``Free-configuration Biased Sampling for Motion Planning'' \cite{bialkowski_IROS13}\\

3: 

4: \noindent The following is the abstract of the original paper:\\

5: 

6: In sampling-based motion planning algorithms the initial step at

7: every iteration is to generate a new sample from the obstacle-free portion of 

8: the configuration space. This is usually accomplished via rejection sampling, 

9: i.e., repeatedly drawing points from the entire space until an obstacle-free 

10: point is found. This strategy is rarely questioned because the extra 

11: work associated with sampling (and then rejecting) useless points contributes at 

12: most a constant factor to the planning algorithm's asymptotic runtime complexity. 

13: However, this constant factor can be quite large in practice.

14: We propose an alternative approach that enables sampling from a 

15: distribution that provably converges to a uniform distribution over {\it only} the obstacle-free space. Our method works by storing empirically observed estimates of obstacle-free space in a point-proximity data structure, and then using this 

16: information to generate future samples. Both theoretical and experimental 

17: results validate our approach.

18: 

19: 

20: \end{abstract}

21: