Abstract
This paper describes a stochastic mapping to understand the situation of an autonomous mobile robot in unknown environment. The proposed stochastic mapping use two-dimensional grid map. In each cell of the grid, Observations use dense stereo vision to measure the variance of the heights. The proposed mapping maintains a posterior distribution over the height variance in each cell. The distribution is used to calculate the likelihood of each observation by effectual statistical processing. This statistical processing is to update the sufficient statistics of a gamma distribution over the precision of heights in each cell. To update each cell is a map update step in Gamma SLAM. By the mapping technique, the grid map has less observation noise. We demonstrate performance on real outdoor environment, and show that the proposed mapping is effective against noisy observations.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
