Rosemarie Velik

Hierarchical Semantic Processing Architecture for Smart Sensors in Surveillance Networks

Data acquisition by multidomain data acquisition provides means for environment perception usable for detecting unusual and possibly dangerous situations. When being automated, this approach can simplify surveillance tasks required in, for example, airports or other security sensitive infrastructures. This paper describes a novel architecture for surveillance networks based on combining multimodal sensor information. Compared to previous methodologies using only video information, the proposed approach also uses audio data thus increasing its ability to obtain valuable information about the sensed environment. A hierarchical processing architecture for observation and surveillance systems is proposed, which recognizes a set of predefined behaviors and learns about normal behaviors. Deviations from “normality” are reported in a way understandable even for staff without special training. The processing architecture, including the physical sensor nodes, is called smart embedded network of sensing entities (SENSE).

Behavior Learning in Dwelling Environments With Hidden Markov Models

Building automation systems (BASs) have seen widespread distribution also in private residences over the past few years. The ongoing technological developments in the fields of sensors, actuators, as well as embedded systems lead to more and more complex and larger systems. These systems allow ever-better observations of activities in buildings with a rapidly growing number of possible applications. Unfortunately, control systems with lots of parameters, which would be normally utilized, are hard to describe and-from a context-deriving view-hard to understand with standard control engineering techniques. This paper presents an approach to how statistical methods can be applied to (future) BASs to extract semantic and context information from sensor data. A hierarchical model structure based on hidden Markov models is proposed to establish a framework. The lower levels of the model structure are used to observe the sensor values themselves, whereas the higher levels provide a basis for the semantic interpretation of what is happening in the building. Ultimately, the system should be able to give a condensed overview of the daily routine of a sensor or the process that the sensor observes. While knowing the context of the sensor, a human operator can easily interpret the result.

Autonomous Perception and Decision Making in Building Automation

System complexity has reached a level where it is hard to apply existing information analysis methods to automatically derive appropriate decisions. Building automation is on the verge of being unable to extract relevant information and control a building accordingly. Many different industries in todays automation could provide information by means of different sensors, but the ability to integrate this information is missing. This paper describes approaches on how to cope with increased complexity by introducing models for perception and decision making that are based on findings in neuroscience and psychoanalysis, scientific disciplines that are far-off from engineering but nevertheless promise valuable contributions to intelligent automation.

Episodic memory for autonomous agents

Knowledge about already experienced situations improves the quality of decision making. The outcomes of decisions in the past can be evaluated and used as a feedback mechanism. This enables the decision making unit is able to learn from past situations. In neuro-psychanalysis, this kind of data storage is called episodic memory. This work describes an episodic memory architecture derived from theories from psychology and neuro-psychoanalysis. The intended applications for this architecture are autonomous agents and building automation systems. The results of the simulation with the autonomous agents have shown that agents with the episodic memory adapt faster to new situations than memory-less agents.

High-level hierarchical semantic processing framework for smart sensor networks

This paper presents the framework of a novel approach to combine multi-modal sensor information from audio and video modalities to gain valuable supplementary information compared to traditional video-based observation systems or even just CCTV systems. A hierarchical, multi-modal sensor processing architecture for observation and surveillance systems is proposed. It recognizes a set of pre-defined behavior and learns about usual behavior. Deviations from ldquonormalityrdquo are reported in a way understandable even for staff without special training. The processing architecture including the physical sensor nodes is called SENSE (smart embedded network of sensing entities) (Zucker and Frangu, 2007).

RETRACTED: From single neuron-firing to consciousness—Towards the true solution of the binding problem

The human brain consists of millions of neural nerve cells being interconnected and firing in parallel in order to process information. A fundamental question is how this parallel neuron-firing can result in a unified experience. This is the so-called binding problem--a problem that is one of todays key questions about brain function and that has puzzled researchers for decades. This article gives a review about the last 50 years of research in this area. It explains what the binding problem is, what classes of binding problems exist, and what the potential solutions suggested so far look like.

Emulating the perceptual system of the brain for the purpose of sensor fusion

This work presents a bionic model derived from research findings about the perceptual system of the human brain to build next generation intelligent sensor fusion systems. Therefore, a new information processing principle called neuro-symbolic information processing is introduced. According to this method, sensory data are processed by so-called neuro-symbolic networks. The basic processing units of neuro-symbolic networks are neuro-symbols. Correlations between neuro-symbols of a neuro-symbolic network can be learned from examples. Perception is based on sensor data as well as on interaction with cognitive processes like focus of attention, memory, and knowledge. Additionally, a mechanism for evaluating perception by emotions is suggested.

Scenario recognition in modern building automation

Abstract Modern building automation has to deal with very different types of demands, depending on the use of the building and therefore the persons acting within this building. To meet the demands of situation awareness in modern building automation, scenario recognition becomes more and more important to detect such demands and react to them. Two concepts of scenario recognition and their implementation will be introduced, one based on predefined templates and the other using an unsupervised learning algorithm using statistical methods. Implemented applications will be described and their advantages and disadvantages outlined. Copyright

Neuro-symbolic networks: introduction to a new information processing principle

Neural networks and symbolic systems are two different approaches to model cognitive functions. Both methods have certain strengths but also suffer from a number of weak points, which are however disparate. By combining characteristics of neural and symbolic processing, these weaknesses could be overcome. This paper introduces a new information processing principle based on so-called neuro-symbolic networks, which incorporates these two approaches. The utility of the suggested principle is outlined for applications in machine perception.

A model for multimodal humanlike perception based on modular hierarchical symbolic information processing, knowledge integration, and learning

Automatic surveillance systems as well as autonomous robots are technical systems which would profit from the ability of humanlike perception for effective, efficient, and flexible operation. In this article, a model for humanlike perception is introduced based on hierarchical modular fusion of multi-sensory data, symbolic information processing, integration of knowledge and memory, and learning. The model is inspired by findings from neuroscience. Information from diverse sensors is transformed into symbolic representations and processed in parallel in a modular, hierarchical fashion. Higher-level symbolic information is gained by combination of lower-level symbols. Feedbacks from higher levels to lower levels are possible. Relations between symbols can be learned from examples. Stored knowledge influences the activation of symbols. The model and the underlying concepts are explained by means of a concrete example taken from building automation.