This chapter sets out to explore the intricacies behind developing a hybrid system for real-time autonomous robot navigation, with target pursuit and obstacle avoidance behaviour, in a dynamic environment. Three complete systems are described, namely, a cascade of four fuzzy systems, a hybrid fuzzy A* system, and a hybrid fuzzy A* with a Voronoi diagram. A highly reconfigurable integration architecture is presented, allowing for the harmonious interplay between the different component algorithms, with the option of engaging or disengaging from the system. The utilization of both global and local information about the environment is examined, as well as an additional optimal global path-planning layer. Moreover, how a fuzzy system design approach could take advantage of the presence of symmetry in the input space, cutting down the number of rules and membership functions, without sacrificing control precision is illustrated. The efficiency of all the algorithms is demonstrated by employing them in a simulation of a real-world system: the robot soccer game. Results indicate that the hybrid system can generate smooth, near-shortest paths, as well as near-shortest-safest paths, when all component algorithms are activated. A systematic approach to calibrating the system is also provided.
|Title of host publication||Efficiency and Scalability Methods for Computational Intellect|
|Editors||Boris Igelnik, Jacek M. Zurada|
|Number of pages||27|
|ISBN (Print)||1466639423, 9781466639423|
|Publication status||Published - 30 Apr 2013|