This is done by expending each obstacle, i.e. C-spaceīefore the robot uses the map for planning the grid map will typically be processed to create a so called C-space (configuration-space) map. If the resolution is 5cm then one cell in a 2D grid covers a 25cm² area. The resolution describes how big of an area each cell covers. We will work with a 2D occupancy grid but 3D grid maps are often used as well. More cells means that it is possible to map a larger area.
The occupancy grid is characterized by the number of cells and the resolution. The value of the cell can tell us if the area is free, occupied, unknown, or something else. Here the world is represented as a grid, where each cell of the grid corresponds to an area in the world. Occupancy grid mapping is one of many mapping algorithms.
Therefore it is of great benefit if the robot can construct a map by itself from scratch. Even if there are maps available, such as blueprints for a building, they are not always useful for the robot and might be incorrect (too old, building collapsed, etc). In many situations we cannot assume that the robot can be given a map in advance. Autonomous robots can use maps in a number of different ways, search and rescue robots can use them to make sure they search through the whole building instead of moving in between the same rooms over and over, autonomous cars use them in order to find a path that leads to the desired location, multicopters can use the maps to localize themself in order to stay in the air. Mapping is one of the core competencies of truly autonomous robots.
0 kommentar(er)
