Development of a Hybrid Architecture for the Autonomous Control of Mobile Robots

One of the most significant challenges in mobile robotics are found in the autonomous navigation area, where the robot must be able to feel and act in an uncertain and geometrically restricted environment, without suffering any type of external interference. The development of an architecture that allows the autonomous navigation arises as a unique combination of issues such as perception, localisation, mapping, path-planning and movement performance. The objective of this work is the development of a hybrid architecture (coordinating technics of planning and reaction) for the autonomous navigation, allowing a mobile robot to navigate in a known and structured environment. This architecture is being developed in a modular and incremental way allowing the incorporation of several techniques of mapping, localisation and path-planning independently of the reactive strategy used by the robot. The proposed arquitecture aims to integrate a module of path-planning using Voronoi diagram and a module of reactive behavior using fuzzy controllers for the control of direction and speed through a module of coordination. The path-planning shall be implemented using a road map created a priori and the Dijkstra’s Algorithm will be used to search the best path. After the generation of the path to be followed, the module of coordination will be responsible for sending actions to control the robot using the sensory information to determine if the robot will follow the path planned or will execute reactive actions. In addition, the module of coordination monitors the execution of the defined path, using a fuzzy controller, deciding if the robot achieved the goal or whether it is necessary to re-planning the path. The proposed arquitecture is being validated in a robot Nomad XR4000. The first results of this work are: the graphic interface developed to integrate the modules and a prototype of the planning module.