| 1. |
|
|
| 2. |
|
|
| 3. |
- Heintz, Fredrik, 1975-, et al.
(författare)
-
Bridging the Sense-Reasoning Gap : DyKnow - A Middleware Component for Knowledge Processing
- 2008
-
Ingår i: IROS Workshop on Current Software Frameworks in Cognitive Robotics Integrating Different Computational Paradigms.
-
Konferensbidrag (refereegranskat)abstract
- Developing autonomous agents displaying rational and goal-directed behavior in a dynamic physical environment requires the integration of both sensing and reasoning components. Due to the different characteristics of these components there is a gap between sensing and reasoning. We believe that this gap can not be bridged in a single step with a single technique. Instead, it requires a more general approach to integrating components on many different levels of abstraction and organizing them in a structured and principled manner. In this paper we propose knowledge processing middleware as a systematic approach for organizing such processing. Desirable properties of such middleware are presented and motivated. We then go on to argue that a declarative streambased system is appropriate to provide the desired functionality. Finally, DyKnow, a concrete example of stream-based knowledge processing middleware that can be used to bridge the sense-reasoning gap, is presented. Different types of knowledge processes and components of the middleware are described and motivated in the context of a UAV traffic monitoring application.
|
|
| 4. |
- Heintz, Fredrik, 1975-, et al.
(författare)
-
Bridging the Sense-Reasoning Gap Using DyKnow : A Knowledge Processing Middleware Framework
- 2007
-
Ingår i: Proceedings of the 30th Annual German Conference on Artificial Intelligence (KI). - Berlin, Heidelberg : Springer. - 9783540745648 ; , s. 460-463
-
Konferensbidrag (refereegranskat)abstract
- To achieve complex missions an autonomous unmanned aerial vehicle (UAV) operating in dynamic environments must have and maintain situational awareness. This can be achieved by continually gathering information from many sources, selecting the relevant information for current tasks, and deriving models about the environment and the UAV itself. It is often the case models suitable for traditional control, are not sufficient for deliberation. The need for more abstract models creates a sense-reasoning gap. This paper presents DyKnow, a knowledge processing middleware framework, and shows how it supports bridging the gap in a concrete UAV traffic monitoring application. In the example, sequences of color and thermal images are used to construct and maintain qualitative object structures. They model the parts of the environment necessary to recognize traffic behavior of tracked vehicles in real-time. The system has been implemented and tested in simulation and on data collected during flight tests.
|
|
| 5. |
- Heintz, Fredrik, 1975-
(författare)
-
DyKnow : A Stream-Based Knowledge Processing Middleware Framework
- 2009
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- As robotic systems become more and more advanced the need to integrate existing deliberative functionalities such as chronicle recognition, motion planning, task planning, and execution monitoring increases. To integrate such functionalities into a coherent system it is necessary to reconcile the different formalisms used by the functionalities to represent information and knowledge about the world. To construct and integrate these representations and maintain a correlation between them and the environment it is necessary to extract information and knowledge from data collected by sensors. However, deliberative functionalities tend to assume symbolic and crisp knowledge about the current state of the world while the information extracted from sensors often is noisy and incomplete quantitative data on a much lower level of abstraction. There is a wide gap between the information about the world normally acquired through sensing and the information that is assumed to be available for reasoning about the world.As physical autonomous systems grow in scope and complexity, bridging the gap in an ad-hoc manner becomes impractical and inefficient. Instead a principled and systematic approach to closing the sensereasoning gap is needed. At the same time, a systematic solution has to be sufficiently flexible to accommodate a wide range of components with highly varying demands. We therefore introduce the concept of knowledge processing middleware for a principled and systematic software framework for bridging the gap between sensing and reasoning in a physical agent. A set of requirements that all such middleware should satisfy is also described.A stream-based knowledge processing middleware framework called DyKnow is then presented. Due to the need for incremental refinement of information at different levels of abstraction, computations and processes within the stream-based knowledge processing framework are modeled as active and sustained knowledge processes working on and producing streams. DyKnow supports the generation of partial and context dependent stream-based representations of past, current, and potential future states at many levels of abstraction in a timely manner.To show the versatility and utility of DyKnow two symbolic reasoning engines are integrated into Dy-Know. The first reasoning engine is a metric temporal logical progression engine. Its integration is made possible by extending DyKnow with a state generation mechanism to generate state sequences over which temporal logical formulas can be progressed. The second reasoning engine is a chronicle recognition engine for recognizing complex events such as traffic situations. The integration is facilitated by extending DyKnow with support for anchoring symbolic object identifiers to sensor data in order to collect information about physical objects using the available sensors. By integrating these reasoning engines into DyKnow, they can be used by any knowledge processing application. Each integration therefore extends the capability of DyKnow and increases its applicability.To show that DyKnow also has a potential for multi-agent knowledge processing, an extension is presented which allows agents to federate parts of their local DyKnow instances to share information and knowledge.Finally, it is shown how DyKnow provides support for the functionalities on the different levels in the JDL Data Fusion Model, which is the de facto standard functional model for fusion applications. The focus is not on individual fusion techniques, but rather on an infrastructure that permits the use of many different fusion techniques in a unified framework.The main conclusion of this thesis is that the DyKnow knowledge processing middleware framework provides appropriate support for bridging the sense-reasoning gap in a physical agent. This conclusion is drawn from the fact that DyKnow has successfully been used to integrate different reasoning engines into complex unmanned aerial vehicle (UAV) applications and that it satisfies all the stated requirements for knowledge processing middleware to a significant degree.
|
|
| 6. |
- Heintz, Fredrik, 1975-, et al.
(författare)
-
DyKnow Federations : Distributing and Merging Information Among UAVs
- 2008
-
Ingår i: Proceedings of the 11th International Conference on Information Fusion (FUSION). - : IEEE conference proceedings. - 9783800730926
-
Konferensbidrag (refereegranskat)abstract
- As unmanned aerial vehicle (UAV) applications become more complex and versatile there is an increasing need to allow multiple UAVs to cooperate to solve problems which are beyond the capability of each individual UAV. To provide more complete and accurate information about the environment we present a DyKnow federation framework for information integration in multi-node networks of UAVs. A federation is created and maintained using a multiagent delegation framework and allows UAVs to share local information as well as process information from other UAVs as if it were local using the DyKnow knowledge processing middleware framework. The work is presented in the context of a multi UAV traffic monitoring scenario.
|
|
| 7. |
- Heintz, Fredrik, 1975-, et al.
(författare)
-
FlexDx: A Reconfigurable Diagnosis Framework
- 2008
-
Ingår i: Proceedings of the 19th International Workshop on Principles of Diagnosis (DX).
-
Konferensbidrag (refereegranskat)abstract
- Detecting and isolating multiple faults is a computationally intense task which typically consists of computing a set of tests, and then computing the diagnoses based on the test results. This paper describes FlexDx, a reconfigurable diagnosis framework which reduces the computational burden by only running the tests that are currently needed. The method selects tests such that the isolation performance of the diagnostic system is maintained. Special attention is given to the practical issues introduced by a reconfigurable diagnosis framework such as FlexDx. For example, tests are added and removed dynamically, tests are partially performed on historic data, and synchronous and asynchronous processing are combined. To handle these issues FlexDx uses DyKnow, a stream-based knowledge processing middleware framework. The approach is exemplified on a relatively small dynamical system, which still illustrates the computational gain with the proposed approach.
|
|
| 8. |
- Heintz, Fredrik, 1975-, et al.
(författare)
-
From Images to Traffic Behavior - A UAV Tracking and Monitoring Application
- 2007
-
Ingår i: Proceedings of the 10th International Conference on Information Fusion (FUSION). - : IEEE conference proceedings. - 9780662458043 - 9780662478300
-
Konferensbidrag (refereegranskat)abstract
- An implemented system for achieving high level situation awareness about traffic situations in an urban area is described. It takes as input sequences of color and thermal images which are used to construct and maintain qualitative object structures and to recognize the traffic behavior of the tracked vehicles in real time. The system is tested both in simulation and on data collected during test flights. To facilitate the signal to symbol transformation and the easy integration of the streams of data from the sensors with the GIS and the chronicle recognition system, DyKnow, a stream-based knowledge processing middleware, is used. It handles the processing of streams, including the temporal aspects of merging and synchronizing streams, and provides suitable abstractions to allow high level reasoning and narrow the sense reasoning gap.
|
|
| 9. |
- Heintz, Fredrik, 1975-, et al.
(författare)
-
Stream Reasoning in DyKnow: A Knowledge Processing Middleware System
- 2009
-
Ingår i: Proceedings of the Stream Reasoning Workshop. - : M. Jeusfeld c/o Redaktion Sun SITE, Informatik V, RWTH Aachen.
-
Konferensbidrag (refereegranskat)abstract
- The information available to modern autonomous systems is often in the form of streams. As the number of sensors and other stream sources increases there is a growing need for incremental reasoning about the incomplete content of sets of streams in order to draw relevant conclusions and react to new situations as quickly as possible. To act rationally, autonomous agents often depend on high level reasoning components that require crisp, symbolic knowledge about the environment. Extensive processing at many levels of abstraction is required to generate such knowledge from noisy, incomplete and quantitative sensor data. We define knowledge processing middleware as a systematic approach to integrating and organizing such processing, and argue that connecting processing components with streams provides essential support for steady and timely flows of information. DyKnow is a concrete and implemented instantiation of such middleware, providing support for stream reasoning at several levels. First, the formal KPL language allows the specification of streams connecting knowledge processes and the required properties of such streams. Second, chronicle recognition incrementally detects complex events from streams of more primitive events. Third, complex metric temporal formulas can be incrementally evaluated over streams of states. DyKnow and the stream reasoning techniques are described and motivated in the context of a UAV traffic monitoring application.
|
|
| 10. |
- Krysander, Mattias, 1977-, et al.
(författare)
-
Dynamic Test Selection for Reconfigurable Diagnosis
- 2008
-
Ingår i: Proceedings of the 47th IEEE Conference on Decision and Control. - : IEEE. - 9781424431243 - 9781424431236 ; , s. 1066-1072
-
Konferensbidrag (refereegranskat)abstract
- Detecting and isolating multiple faults is a computationally intense task which typically consists of computing a set of tests, and then computing the diagnoses based on the test results. This paper proposes a method to reduce the computational burden by only running the tests that are currently needed, and dynamically starting new tests when the need changes. A main contribution is a method to select tests such that the computational burden is reduced while maintaining the isolation performance of the diagnostic system. Key components in the approach are the test selection algorithm, the test initialization procedures, and a knowledge processing framework that supports the functionality needed. The approach is exemplified on a relatively small dynamical system, which still illustrates the complexity and possible computational gain with the proposed approach.
|
|
