Robotic mapping
From Wikipedia, the free encyclopedia
The problem of Robotic mapping is related to cartography. The goal is for an autonomous robot to be able to construct (or use ) a map or floor plan and to localize itself in it.
Todd et al (1994) have shown that evolutionarily shaped blind action may suffice to keep some animals alive. For some insects for example, the environment is not interpreted as a map, and they survive only with a triggered response.
But a slightly more elaborated navigation strategy dramatically enhances the capabilities of the robot. Cognitive maps (Tolman 1948) enable planning capacities, and use of current perceptions, memorized events, and expected consequences.
A good algorithm in robotic mapping may combine the information from the past, the present and the future (Trullier et al. 1997).
The problem can be decomposed in three processes (Levitt and Lawton 1990 ; Balakrishnan et al. 1999) : map learning, localisation, path-planning.
Contents |
[edit] Available information
The robot has two sources of information: the idiothetic and the allothetic sources.
When it moves, the robot integrates its position by counting the number of wheel turns it has done. This corresponds to the idiothetic source. It can give the absolute position of the robot. But it is subject of cumulative error which can grow fast.
The allothetic source corresponds to the sensors of the robot, like a camera, a microphone, laser, sonar, ... The problem is the "perceptual aliasing". This means that two different places can be perceived the same. For example, in a building, you can't know where you are only with the visual information, because all the corridors look the same.
[edit] Map representation
The internal representation of the map can be "metric" or "topological". The metric framework is the most common for humans and considers a two dimensional space in which it places the objects. The objects are placed with precise coordinates. This representation is very useful, but is sensitive to noise and it is difficult to calculate precisely the distances.
The topological framework only considers places and relations between them. Often, the distances between places are stored. The map is then a graph, in which the nodes corresponds to places and arcs correspond to the paths.
Almost all state-of-the-art solutions use probabilistic representations of the map, in order to correctly handle uncertainty.
There are three main methods of Map representations:
[edit] Free Space Maps
- Spatial Graphs
- Voronoi Diagrams
- Generalised Voronoi Diagrams
[edit] Object Maps
[edit] Composite Maps
- Area Grids
- Point Grids
- Quad Trees
These employ the notion of a grid, but permit the resolution of the grid to vary so that it can become finer where more accuracy is needed and more coarse where the map is uniform.
[edit] Map learning
Map-learning can't be separated from the localization process so it is difficult because errors in localization are incorporated into the map. This problem is commonly referred to as Simultaneous localization and mapping (SLAM).
An important additional problem is to determine whether the robot is in a part of environnement already stored or never visited.
[edit] Path planning
The path planning problem is not an important issue if the map and localization are accurate. For the metric representation of the map, the robot can find short cuts in the map. For the topological representation, the problem of path planning is a classical problem of finding the shortest path between two nodes in a graph.
[edit] Navigation
- Further information: navigation research
Outdoor robots can use GPS in a similar way to automotive navigation systems. Alternative systems can be used with floor plan instead of maps for indoor robots, combined with localization wireless hardware.
[edit] See also
- Automotive navigation system
- Domestic robot and indoor transport.
- Floor plan
- GPS
