Implementation of real time spatial mapping in robotic systems through self-organizing neural networks

A methodology has been developed which allows mobile robots to learn and build maps of its operating environment by relying on its range sensors. The maps, described with respect to the robot's internal frame, were developed in real time by correlating robot position and sensory data. The methodology exploits the principle of Self-Organization, implemented as an artificial neural network module which processes incoming sensor range data. Experiments focusing on indoor applications demonstrated the ability of a robot to build maps of geometrically complex environments. Results showed improvement in that every single sensor data point contributes equally to the location of the neurons of the spatial map at the end of the learning process.