Mobile robot map building from time-of-flight camera

A map building algorithm for mobile robots is introduced in this paper. The perceived environment is represented in a map containing in each cell a probability of presence of an object or part of an object. The environment is represented as a collection of modular occupancy grids which are added to the map as far as the mobile robot finds objects outside the existing grids. In this approach a time-of-flight (ToF) camera is exploited as a range sensor for mapping. Indeed, one of the areas where ToF sensors are adequate is in obstacle avoidance, because the detection region is not only horizontal but also vertical, allowing to detect obstacles with complex shapes. The main steps of the map building algorithm are extensively described in the paper. The results of testing the algorithm are considered in two different indoor environments.