A Hybrid Architecture for Planning and Execution of Multi-Behavior Data Acquisition Missions

This paper addresses the issue of designing integrated deliberative-reactive architectures for multi-behavior robot navigation control. The objective of the study is to devise and investigate a methodology for designing robust planning and control systems equipped with a high level of intelligence and capable of navigating a mobile platform, at a high level of performance, in complex environment conditions, where the mobile robot multi-task operation is subject to different behaviors. A formal model of the integrated architecture is presented. Components of the model incorporate hybrid intelligence techniques, allowing the robot to perform different patterns of behavior for different purposes. Metaheuristic procedures enhance the deliberative level producing the optimal global path and the optimal sub-global path. Multiple search methods are proposed to optimize and enable multi-behavior path planning navigation based on waypoints approach. A behavior selector is employed for controlling and executing the appropriate behavior to perform complex tasks along the global path. On the reactive level, fuzzy behavior-based systems are employed to execute different robot tasks including conflicting behaviors. A navigation behavior control module regulates the relation between the navigation levels and as well as executes control on each navigation component. Although designed for the execution of data acquisition missions, the proposed architecture is general enough to show good performance in a variety of complex conditions. Experimental results obtained by using a Khepera robot demonstrate the validity of the presented hybrid architecture in a critical dynamic and complex environment.