Mobile robot navigation
Mobile robot navigation is the process of reading, and controlling the movement of a mobile robot from one location to another. The main types of robots are manipulators, mobile and biological inspired ones. The most attractable research area in robotics is mobile robots. Several companies are developing the different models of mobile robot like wheeled and human like robots. These robots are used in many applications for examples rescue and search like areas.
Since the presentation of the play Rossum's Univeral Robots in 1920 the word robot entered in the English language . The term robot comes from robota which is a Czech word meaning forced labor. Robot Institute of America defines a robot as "A reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized devices through various programmed motions for the performance of a variety of tasks" .
2.3. Types of Robots
The main types of the robots are
a) Robot Manipulators
b) Mobile Robots
c) Biologically Inspired Robots
2.3.1. Robot Manipulators
A robot manipulator is also called robot arm which are fixed and design to perform some specific task such as welding, painting and palletizing. In the recent year as the technologies are growing the new applications of robot manipulators are also discovered, such as rehabilitation, surgery assistance etc. Robot manipulators are used in different areas such as industrial, medical and rehabilitation robots.
Industrial Robots: These robots were introduced in early 1960 for the production process . Industrial robots are used for any task that involves repetitiveness, accuracy, endurance, speed, and reliability. These tasks done by the humans are now increasingly being done by industrial robots.
Medical Robots:Robots are also occupying the field of medicine, not to replace the qualified person such as doctors but they assist them in routine work and accuracy. A wide variety of medical robots are used in applications such as telesurgery, laboratory robots, surgical training, remote surgery and hospital robots.
Rehabilitation Robots: Activity in this field began in the 1960s  and now the commercially successful products are available. These robots can be used in different applications such as robots for neuro rehabilitation, power augmentation etc.
2.3.2. Mobile Robots
Mobile robot describes any robotic system which is able to perform tasks in different locations and having an ability to move from one location to another called mobility. These robots can perform more task than manipulators and open the new area of research such as navigation problem. Mobile robot was introduced in 1968 like automated guided vehicles and vehicles transporting tools. In this area researchers can deals now with the indoor and outdoor navigation. These robots can be used in different applications like underwater robots, rescue robots, service robots and bomb disposal robots.
Underwater Robots: More than 70% of the earth is covered by ocean. However, a little effort compare to space has been made to utilize this vast resource. During the last decades, the underwater robotic vehicles were rapidly used because these vehicles can be operated in the deeper where the drivers can not reach. Applications of such vehicles include fishing, rescue, waste cleaning, and underwater pollution monitoring.
Rescue Robots: Rescue robots are designed for the purpose of aiding rescue workers. These robots are usually used for exploration and collecting the information of victims in the hazardous environment where it is too risky to send human inside.
2.3.3. Biologically Inspired Robots
Walking Robots: There is a lot of research work for mobile robots that use legs for navigation. The legs of the walking robot are manipulators thus both have some common technical problems. The mobile robots having legs for movement have an advantage over the mobile robots using wheels for movement such as better movement on stairs, obstacles and ditches.
Humanoid Robots: Overall appearance of a robot based on the human body are called humanoid robot.
2.4. Mobile Robot Navigation System
A mobile robot is developed and programmed to accomplish a mission, the robot has to move to the target positions. The problem implies that the robot has to localize the target. It requires a world representation commonly defined as a map, where the robot identifies the target and has to estimate its current position the localization problem. The map could be complete, which addresses the map building problem. Finally, to accomplish the mission, the robot has to move to reach the target goal, selecting the best trajectory, which is the navigation problem.
The navigation is accomplished based on the available map. There are different types of maps, as shown in Fig 1, derived from . Same language is used for navigation and map which means that a metric navigation requires the metric information of the map and topological navigation requires the topological information of the map. A metric navigation or a topological navigation can be used for the same mission, by using different levels of abstraction. For example, when a robot moves from a location A to a location B, can be use any level of navigation. Move from location A to location B is the most abstract level of navigation. Moving through the doors and passageways by metric navigation is a lower level of abstraction. The lowest level of navigation, the metric navigation, includes the path following that requires a metric referential, as illustrated in Fig 1.
In this thesis the navigation of mobile robots with a high level of abstraction are considered and defined as topological navigation. High level of abstraction is possible to define the locations as rooms, doors, passageways and stairs.
There are three main problems in the mobile robot navigation called as localization, navigation and mapping. It is necessary to keep track of the mobile robots location when these are moving in the environment known as localization problem, generate a path to the final locations known as the navigation problem and changes to a new location when robots are moving known as mapping problem. The robot's position is relative to the point when the robot was switched on.
2.5. Environment Representations
One of the key issues in mobile robot navigation is how to represent the environment, which is used to support the mobile robot for navigation. However navigation depends on the adopted method for representation of environment. These methods are classified as topological, geometric and hybrid.
2.5.1. Topological Maps
A Topological Map is the representation of the environment by nodes and links. Nodes and links represent the distinct locations and motion between nodes respectively. A topological map is also a featured based map composed of optimal tracking capabilities of feature maps and scalability of topological map. The scalability of a topological map depends on the type and amount of features used and these features can describe the accuracy for coverage of a large and small physical area. An optical tracking capability is the feature includes more information than the metric location which can reduce the uncertainty perception. Topological approaches do not require the exact determination of the geometric localization of the robot and explore the acquired information efficiently, because information is represented with accuracy level adapted to the needs. A topological map is built by driving a robot through an indoor environment .
2.5.2. Geometric Maps
A Geometric Map is a metric representation of the environment relative to a referential, referential could be local or global. In a geometric map information is referred to physical distances and dimensions of places like walls doors, walls and passageways. Global referential means the reference is equal in all maps and attached to the real world and local referential if it covers a particular place. The geometric maps are commonly based on a 2D/3D referential. Grid maps are included in the class of geometric maps and used for environment representation. These maps consists two or three dimension cells, where each cell contains the information of its occupancy.
2.5.3. Hybrid Maps
Hybrid maps are composition of both topological map and Geometric Maps, aim of this composition is to combining the advantage of both maps. Both approaches have advantages and disadvantages to represent the environment.
Several companies are developing and implementing models of mobile robots such as wheeled robots, developed for testbed, repetitive and dangerous tasks, or simply for entertainment and human like robots. Their movement becomes similar to the humans walking, climbing, and picking something and these mobile robots are equipped with reactive sensors, as touch, force, vision and others. It is most important that the robots and humans can share the same language and operate in the same scenario. These robots not only require the metric information but also require the high level of abstraction. For example given a task, "pick up the chair and put it in the room" for this task metric information to identify the distance to the chair and how to pick the chair and for more important it require information of high level abstraction that where is the room and how to reach that room. The scenarios where the mobile robots can operate cover important issues of mobile robot navigation can be known or unknown environment, in thesis we only focused on mobile robot navigation in known environment.
2.6.1. Rescue and Search like applications
Thousand of situations occur which require emergency response such as rescue, medical involvement, hazardous environment where the human life is in danger. The most serious issues involving large number of heterogeneous robots in unfriendly environment is the disaster rescue, and several researchers are working on this issue.
The most growing number of researchers is working on rescue and search area for robotic systems. University of South Florida established a team under the supervision of Robin Murphy shortly after the Word Trade Center collapsed. They use tele-operated robots to assist the search for survivors and bodies at level zero .
 R. Simmons and S. Koenig, "Probabilistic Robot Navigation in Partially Observable Environments," Proceedings of the International Joint Conference on Artiﬁcial Intelligence, pp. 1080-1087, 1995.
 J.D. Naughton, "Futurology and robots: Karel Caper's R.U.R," Culture, Theory and Critique, Volume 28, Issue 1, pp 72 - 86, 1984.
 B. Gunnar, O. Magnus and C. Per, "Robotic are welding - trends and developments for higher autonomy," Industrial Robot: An International Journal, Volume 29, Issue 2, pp 98 - 104, 1984.
 Y. Kim and A.M. Cook, Manipulation and Mobility Aids, In J.G.Webster et al, Eds. Electronic Devices for Rehabilitation. London, U.K. Chapman and Hall, 1985.
 G.L.Dudek, "Environment representation using multiple abstraction levels," Proceedings of the IEEE, Volume 84, Issue 11,pp. 1684 - 1704, 1996.
 J. Gaspar, N.Winters and J. Santos "Victor. Vision based navigation and environmental representations with an omni-directional camera. IEEE TRA, December 200.
 J. Casper and R. Murphy "Human-robot interactions during the robot-assisted urban search and rescue response at the world trade center," IEEE Transactions on Systems, Man and Cybernetics Part B, pp. 367 - 385, June 2003.