Artificial intelligence and mobile robots
An Behavior-based Robotics
A hybrid architecture for autonomous navigation using a CBR reactive layer
IAT '03 Proceedings of the IEEE/WIC International Conference on Intelligent Agent Technology
Control of a differentially driven mobile robot using radial basis function based neural networks
WSEAS Transactions on Systems and Control
| Hi-index | 0.00 |
Operating environments for autonomous mobile robots are characterized by short-term and long-term changes or moving obstacles that make it very difficult for the robot to acquire and maintain an exact model of the world. For robots to operate successfully under such conditions, they have to use any knowledge that is available. In order to continuously perceive information about the environment and update its model of the world, an autonomous robot is typically equipped with several sensor systems. In this paper, we present a control architecture where perceptions are also capable of directly activating robot behaviour. The control architecture scheme follows the hybrid guidelines and it maintains an environment representation where the highest level is built from perception outcomes. These perceptions are organized into a set of modules that perform different levels of sensory representation: primitives, such as localization or landmark detection, and compound perceptions, such as feature maps (local SLAM). The low reactive layer of the architecture is based on a set of behaviours which only provide the robot with navigation capabilities. Finally, the deliberative layer builds a symbolic representation of the environment (topological map) and integrates suitable algorithms to accomplish the execution of the task.