We present a realistic neural network - the canonical cortical module - built on basic principles of cortical organization. These principles are: opponent cells principle, assembly principle, canonical cortical circui...
详细信息
We present a realistic neural network - the canonical cortical module - built on basic principles of cortical organization. These principles are: opponent cells principle, assembly principle, canonical cortical circuit principle and modular principle. When applied to visual images, the network explains orientational and spatial frequency filtering functions of neurons in the striate cortex. Two patterns of joint distribution of opponent cells in the inhibitory cortical layer are presented: pinwheel and circular. These two patterns provide two Gestalt descriptions of local (within the frames of one module) visual image: circle-ness and cross-ness. These modules were shown to have a power for shape detection and texture discrimination. They also provide an enhancement of signal-to-noise ratio of input images. Being modality independent, the canonical cortical module seems to be good tool for bio-fusion for intelligent system control.
Measurement and signal intelligence (MASINT) of the battlespace has created new requirements in information management, communication and interoperability as they effect surveillance and situational awareness. In many...
详细信息
ISBN:
(纸本)0819431877
Measurement and signal intelligence (MASINT) of the battlespace has created new requirements in information management, communication and interoperability as they effect surveillance and situational awareness. In many situations, stand-off remote-sensing and hazard-interdiction techniques over realistic operational areas are often impractical and difficult to characterize. An alternative approach is to implement adaptive remote-sensing techniques with swarms of mobile agents employing collective behavior for optimization of mapping signatures and positional orientation (registration). We have expanded intelligent control theory using physics-based collective behavior models and genetic algorithms to produce a uniquely powerful implementation of distributed ground-based measurement incorporating both local collective behavior, and inter-operative global optimization for sensorfusion and mission oversight. By using a layered hierarchical control architecture to orchestrate adaptive reconfiguration of semi-autonomous robotic agents, we can improve overall robustness and functionality in dynamic tactical environments without information bottlenecking. In our concept, each sensor is equipped with a miniaturized optical reflectance modulator which is interactively monitored as a remote transponder using a laser communication protocol from a remote mothership or operative. Robot data-sharing at the ground level can be leveraged with global evaluation criteria, including terrain overlays and remote imaging data. Information sharing and distributed intelligence opens up a new class of remote sensing applications in which small single-function autonomous observers at the local level can collectively optimize and measure large scale ground-level signatures. As the need for coverage and the number of agents grows to improve spatial resolution, cooperative behavior orchestrated by a global situational awareness umbrella will be an essential ingredient to offset increasing bandw
暂无评论