Andrey Belkin

World modeling for autonomous systems

This contribution proposes a universal, intelligent information storage and management system for autonomous systems, e. g., robots. The proposed system uses a three pillar information architecture consisting of three distinct components: prior knowledge, environment model, and real world. In the center of the architecture, the environment model is situated, which constitutes the fusion target for prior knowledge and sensory information from the real world. The environment model is object oriented and comprehensively models the relevant world of the autonomous system, acting as an information hub for sensors (information sources) and cognitive processes (information sinks). It features mechanisms for information exchange with the other two components. A main characteristic of the system is that it models uncertainties by probabilities, which are handled by a Bayesian framework including instantiation, deletion and update procedures. The information can be accessed on different abstraction levels, as required. For ensuring validity, consistence, relevance and actuality, information check and handling mechanisms are provided.

Data association in a world model for autonomous systems

This contribution introduces a three pillar information storage and management system for modeling the environment of autonomous systems. The main characteristics is the separation of prior knowledge, environment model and sensor information. In the center of the system is the environment model, which provides the autonomous system with information about the current state of the environment. It consists of instances with attributes and relations as virtual substitutes of entities (persons and objects) of the real world. Important features are the representation of uncertain information by means of Degree-of-Belief (DoB) distributions, the information exchange between the three pillars as well as creation, deletion and update of instances, attributes and relations in the environment model. In this work, a Bayesian method for fusing new observations to the environment model is introduced. For this purpose, a Bayesian data association method is derived. The main question answered here is the observation-to-instance mapping and the decision mechanisms for creating a new instance or updating already existing instances in the environment model.

Pre-production validation of the ATLAS Level-1 calorimeter trigger system

The Level-1 calorimeter trigger is a major part of the first stage of event selection for the ATLAS experiment at the LHC. It is a digital, pipelined system with several stages of processing, largely based on FPGAs, which perform programmable algorithms in parallel with a fixed latency to process about 300 Gbyte/s of input data. The real-time output consists of counts of different types of trigger objects and energy sums. Prototypes of all module types have been undergoing intensive testing before final production during 2005. Verification of their correct operation has been performed stand-alone and in the ATLAS test-beam at CERN. Results from these investigations will be presented, along with a description of the methodology used to perform the tests

Prior knowledge employment based on the k-l and tanimoto distances matching for intelligent autonomous robots

Modern autonomous robots are performing complex tasks in a real dynamic environment. This requires real-time reactive and pro-active handling of arising situations. A basis for such situation awareness and handling can be a world modeling subsystem that acquires information from sensors, fuses it into existing world description and delivers the required information to all other robot subsystems. Since sensory information is affected by uncertainty and lacks for semantic meaning, the employment of a predefined information, that contains concepts and descriptions of the surrounding world, is crucial. This employment implies matching of the world model information to prior knowledge and subsequent complementing of the dynamic descriptions with semantic meaning and missing attributes. The following contribution describes a matching mechanism based on the Kullback-Leibler and Tanimoto distances and direct assignment of the prior knowledge for the model complementation.

Beam test of the ATLAS level-1 calorimeter trigger system

The level-1 calorimeter trigger consists of a preprocessor (PP), a cluster processor (CP), and a jet/energy-sum processor (JEP). The CP and JEP receive digitised trigger-tower data from the preprocessor and produce regions-of-interest (RoIs) and trigger multiplicities. The latter are sent in real time to the central trigger processor (CTP) where the level-1 decision is made. On receipt of a level-1 accept, readout driver modules (RODs) provide intermediate results to the data acquisition (DAQ) system for monitoring and diagnostic purposes. RoI information is sent to the RoI builder (RoIB) to help reduce the amount of data required for the level-2 trigger. The level-1 calorimeter trigger system at the test beam consisted of 1 preprocessor module, 1 cluster processor module, 1 jet/energy module and 2 common merger modules. Calorimeter energies were successfully handled throughout the chain and trigger objects sent to the CTP. Level-1 accepts were successfully produced and used to drive the readout path. Online diagnostics were made using 4 RODs. Energy histograms were plotted and the integrity of data between the different modules was checked. All ATLAS detectors in the test beam were able to build full events based on triggers delivered by the calorimeter trigger system.

Three pillar information management system for modeling the environment of autonomous systems

This contribution is about an information management and storage system for modeling the environment of autonomous systems. The three pillars of the system consist of prior knowledge, environment model and sensory information. The main pillar is the environment model, which supplies the autonomous system with relevant information about its current environment. For this purpose, an abstract representation of the real world is created, where instances with attributes and relations serve as virtual substitutes of entities (persons and objects) of the real world. The environment model is created based on sensory information about the real world. The gathered sensory information is typically uncertain in a stochastic sense and is represented in the environment model by means of Degree-of-Belief (DoB) distributions. The prior knowledge contains all relevant background knowledge (e. g., concepts organized in ontologies) for creating and maintaining the environment model. The concept of the three pillar information system has previously been published. Therefore this contribution focuses on further central properties of the system. Furthermore, the development status and possible applications as well as evaluation scenarios are discussed.

Pre-production validation of the ATLAS level-1 calorimeter trigger system

The Level-1 Calorimeter Trigger is a major part of the first stage of event selection for the ATLAS experiment at the LHC. It is a digital, pipelined system with several stages of processing, largely based on FPGAs, which perform programmable algorithms in parallel with a fixed latency to process about 300 Gbyte/s of input data. The real-time output consists of counts of different types of trigger objects and energy sums. Prototypes of all module types have been undergoing intensive testing before final production during 2005. Verification of their correct operation has been performed stand-alone and in the ATLAS test-beam at CERN. Results from these investigations will be presented, along with a description of the methodology used to perform the tests.

On weak distance between distributions in application to tracking

In this paper, we consider the general problem of assessing accuracy losses associated with converting distributions from one representation to the other. Based on distribution theory, we argue that any such quality metric is intrinsically problem-specific, and that the choice of the so-called probe functions is unavoidable. We discuss the meaning of these definitions in the context of tracking, and how probe functions may encode valuable a priori assumptions about sensors and the tracking quality. Based on these ideas, we suggest two novel algorithms: one to prune Gaussian mixtures (GMs) and the other to perform a weighted sampling of GMs. Finally, we compare the tracking quality between identical trackers where GM pruning is done with the suggested and the conventional algorithms.