Path recognizing for Robots through low cost sensors: From Mobile and Humanoid Robot perception model up to simulation of several heuristic approaches for path planning

Navigation of mobile robots is a complex process including perception, localization, path- planning and motion control. Each of those tasks should be completely understand and from robot correct executed in order to navigate successfully. Understanding and interpreting of real working environment by robots is done through perception process. For this purpose, data obtained from multiple sensors should be correctly extracted and interpreted by the robot, especially using the sensor fusion. Once the working environment is correct interpreted, the robot should correctly localize himself. Before a robot moves from any start position to goal position, a path should be planned. This planning of path should include not only obstacle avoiding but also the optimality. Different algorithms for this purpose exist, from road-maps up to heuristic and genetic algorithms. In this work, a new cost oriented method is proposed, which take in consideration Localization and Path Planning. The proposed approaches for path planning are based on heuristic algorithms. The main advantages of such approaches are the time reducing of map-calculating and replacement of expensive high-tech devices.