Fuzzy Logic Based Navigation of Mobile Robots

Abstract

Robots are no longer confined to engineered, well protected sealed corners, but they are currently “employed” in places closer and closer to “us”. Robots are getting out of factories and are finding their way into our homes and to populated places such as, museum halls, office buildings, schools, airports, shopping malls and hospitals. The gained benefit of the potential service and personal robots comes along with the necessity to design the robot in a way that makes it safe for it to interact with humans and in a way that makes it able to respond to a list of complex situations. This includes at least the possibility to have the robot situated in an unknown, unstructured and dynamic environment and to navigate its way in such an environment. One of the fundamental issues to be addressed in autonomous robotic system is the ability to move without collision. An intelligent robot should avoid undesirable and potentially dangerous impact with objects in its environment. This simple capability has been the subject of interest in robotic research. Behavior based navigation systems (Arkin, 1987, 1989; Arkin & Balch, 1997; AlYahmedi et al., 2009; Brooks, 1986, 1989; Fatmi et al. 2006 and Ching-Chih et al. 2010) have been developed as an alternative to the more traditional strategy of constructing representation of the world and then reasoning prior to acting. The main idea of behavior based navigation is to identify different responses (behaviors) to sensory inputs. For example, a behavior could be avoiding obstacles in which sonar information about a close obstacle should result in a movement away from the obstacle. A given set of behaviors is then blended in a certain way to produce either a trade off behavior or a more complex behavior. However, a number of issues with regard to behavior based navigation are still under investigation. These issues range from questions concerning the design of individual behaviors to behavior coordination issues, to intelligently improve “behaviors” through learning. An important problem in autonomous navigation is the need to deal with the large amount of uncertainties of the sensory information received by the robot which is incomplete and approximate as well as with the fact that the environment in which such robots operate contains dynamics and variability elements. A fuzzy logic behavior based navigation approach is introduced in this chapter in order to deal with the uncertainty and ambiguity of the information the system receives. Issues of individual behavior design and action coordination of the behaviors will be addressed using fuzzy logic. The approach described herein, consists of the following four tasks,