Less recent publications of Joachim Hertzberg

	Institute of Computer Science — Knowledge-Based Systems Research Group
	Less recent publications of Joachim Hertzberg, comprehensive list

... until about 2000, by category, including abstracts. Papers with German titles are--unsurprisingly--in German. Some of the more recent papers are directly available as pdf or postscript (ps.gz) sources. If you wish to obtain a paper for which no such source is given, please contact me. A condensed, but recent list can be found at this link.

New (or as yet unpublished) papers
Books
Journal papers
Invited papers, book chapters
Major conference papers

New papers

Paper:: F. Schönherr, M. Cistelecan, J. Hertzberg, Th. Christaller: Extracting Situation Facts from Activation Value Histories in Behavior-Based Robots. To appear in: KI-2001: 24th German / 9th Austrian Conference on Artificial Intelligence (September 19-21, 2001 Vienna, Austria)
Abstract:: The paper presents a new technique for extracting symbolic ground facts out of the sensor data stream in autonomous robots for use under hybrid control architectures, which comprise a behavior-based and a deliberative part. The sensor data are used in the form of time series curves of behavior activation values. Recurring patterns in individual behavior activation curves are aggregated to well-defined patterns, like edges and levels, called qualitative activations. Sets of qualitative activations for different behaviors occurring in the same interval of time are summed to activation gestalts. Sequences of activation gestalts are used for defining chronicles, the recognition of which establishes evidence for the validity of ground facts. The approach in general is described, and examples for a particular behavior-based robot control framework in simulation are presented and discussed.
Full paper in PDF (129 kB)

Books

Also included here are books that I have only (co-)edited rather than written.

Book:: J. Sauer, A. Günter, and J. Hertzberg (eds.): Beiträge zum 10. Workshop ``Planen und Konfigurieren''. infix-Verlag (Proceedings in AI vol. 3), Sankt Augustin, 1996

Book:: H.W. Güsgen and J. Hertzberg: A Perspective of Constraint Based Reasoning. Berlin (Springer, LNAI vol. 597) 1992

Book:: J. Hertzberg (ed.): European Workshop on Planning. EWSP '91, Sankt Augustin, FRG, March 1991, Proceedings. Springer Verlag (Lecture Notes in AI vol. 522), 1991

Book:: J. Hertzberg: Planen. Einführung in die Planerstellungsmethoden der Künstlichen Intelligenz. Mannheim (BI Wissenschaftsverlag), 1989

Journal papers

Paper:: E. Rome, J. Hertzberg, Th. Christaller, F. Kirchner, and U. Licht: Towards Autonomous Sewer Robots: The MAKRO Project. in: J. Urban Water 1:57-70, 1999
Abstract:: We expose current research and development in information technology that deals with building and controlling autonomous service robots for performing inspection tasks in sewerage systems that are inaccessible for humans. The problem is explained, the physical and legal boundary conditions for operating sewer robots are described, the state of current sewer inspection technology is sketched, and the need for using advanced technology in this area motivated. Respective work that we have been doing for the last four years, and that is currently being pursued in the MAKRO project, is presented. The main technological and control problems are described, and how Artificial Intelligence research and technology may be employed to solve the latter.

Paper: F. Kirchner and J. Hertzberg: A Prototype Study of an Autonomous Robot Platform for Sewerage System Maintenance. in: J. Autonomous Robots 4(4):319-331, 1997
Abstract:: The paper describes KURT, an autonomous robot platform prototype that is able to navigate through a network of sewage pipes. KURT is a six-wheeled vehicle; it has modular, layered hardware and control architectures. Its sensor configuration consists mainly of stationary and one flexible ultrasound transducers and of two inclinometers. In experiments run in a test sewerage system consisting of concrete pipes of 600mm diameter, it has proven its abilities to travel safely through straight pipes, to recognize different types of pipe crossings, to turn at crossings, and to navigate from a given start point to arbitrary specified goal points.
Postscript Source (ps.gz, 1700 KB)

Paper:: K. Berns, Th. Christaller, R. Dillmann, J. Hertzberg, W. Ilg, M. Kemmann, E. Rome, and H. Stapelfeldt: LAOKOON -- Lernfähige autonome kooperierende Kanalroboter. in: KI 11(2):28-32, 1997
Abstract:: Die Machbarkeitsuntersuchung LAOKOON hatte zum Ziel, die Möglichkeiten und das Potential des Einsatzes autonomer mobiler Roboter in Abwasserkanälen aufzuzeigen. Der Text skizziert Anwendungsproblem, Ziel und Form der Studie und benennt ihre Hauptergebnisse.
Postscript Source (ps.gz)

Paper:: H.W. Güsgen and J. Hertzberg: Spatial Persistence. J. Appl. Intell. (Special Issue on Spatial and Temporal Reasoning) 6(1):11-28, 1996
Abstract:: We examine the spatial version of the persistence problem. In temporal reasoning, this is the problem of determining whether or not the validity of a fact at some point in time persists until another point in time, given that certain events or processes may happen in between. We show that its analog does intuitively exist in spatial reasoning, and review under the aspect of transferability to space different approaches for achieving persistence in {\it temporal\/} reasoning. Finally, we present reasoning with generalized spatial Allen relations as an instance of reasoning under the assumption of spatial persistence.
Postscript Source (ps.gz, 227 KB)

Paper:: J. Hertzberg and S. Thiébaux: Turning an Action Formalism into a Planner---A Case Study. J. Logic and Computation 4(5):617-654, 1994. (Short version in Proc. ISMIS'94.)
Abstract:: The paper describes a case study that explores the idea of building a planner with a neat semantics of the plans it produces, by choosing some action formalism that is ``ideal'' for the planning application and building the planner accordingly. In general---and particularly so for the action formalism used in this study, which is quite expressive---this strategy is unlikely to yield fast and efficient planners if the formalism is used naïvely. Therefore, we adopt the idea that the planner approximates the theoretically ideal plans, where the approximation gets closer, the more run time the planner is allowed. As the particular formalism underlying our study allows a significant degree of uncertainty to be modeled and copes with the ramification problem, we end up in a planner that is functionally comparable to modern anytime uncertainty planners, yet is based on a neat formal semantics.

Paper:: S. Thiébaux, J. Hertzberg, W. Shoaff, and M. Schneider: A Stochastic Model of Actions and Plans for Anytime Planning under Uncertainty. Int. J. of Intelligent Systems 9, 1994. (Short version in Proc. European Workshop on Planning 1993 (EWSP'93).)
Abstract:: Building planning systems that operate in real domains requires coping with both uncertainty and time pressure. This paper describes a model of reaction plans, which are generated using a formalization of actions and of state descriptions in probabilistic logic, as a basis for anytime planning under uncertainty. The model has the following main features. At the action level, we handle incomplete and ambiguous domain information, and reason about alternative action effects whose probabilities are given. On this basis, we generate reaction plans that specify different courses of action, reflecting the domain uncertainty and alternative action effects; if generation time was insufficient, these plans may be left unfinished, but they can be reused, incrementally improved, and finished later. At the planning level, we develop a framework for measuring the quality of plans that takes domain uncertainty and probabilistic information into account using Markov chain theory; based on this framework, one can design anytime algorithms focusing on those parts of an unfinished plan first, whose completion promises the most ``gain''. Finally, the plan quality can be updated during execution, according to additional information acquired.

Paper:: G. Brewka and J. Hertzberg: How to Do Things with Worlds: On Formalizing Actions and Plans. J. Logic and Computation 3(5):517-532, 1993
Abstract:: The paper begins with a critique of the Ginsberg/Smith approach to reasoning about action which is based on the notion of maximal consistent subsets of world descriptions. There are some deep diffculties with this approach due to a misconception of what is possible and what is not. Incomplete information about the actual state of the world may lead to counterintuitive conclusions about what is true after an action is performed. These problems were first discussed by Winslett, who presents an alternative formalization based on possible models. Her formalization solves problems with incomplete descriptions of the world. However, related problems caused by using disjunctive postconditions to model ambiguous actions remain unsolved. Moreover, Winslett does not take the effects of causality into account when determining the closeness of worlds. We present a novel formalization of actions that avoids these problems.

Paper:: H.W. Guesgen and J. Hertzberg: A Constraint-Based Approach to Spatiotemporal Reasoning. J. Appl. Intell. 3:71-90, 1993
Abstract:: We introduce a form of spatiotemporal reasoning that uses homogeneous representations of time and the three dimensions of space. The basis of our approach is Allen's temporal logic on the one hand and general constraint satisfaction algorithms on the other, where we present a new view of constraint reasoning to cope with the affordances of spatiotemporal reasoning as introduced here. As a realization for constraint reasoning, we suggest a massively parallel implementation in form of Boltzmann machines.

Paper:: E. Rutten and J. Hertzberg: Temporal Planner = Nonlinear Planner + Time Map Manager. AI Communications 6(1):18-26, 1993
Abstract:: We describe how you end up in a temporal planner by building a classical (non-temporal) nonlinear planner on top of a time map management system. In particular, we motivate some requirements for the underlying time map manager, and describe the distribution of labour between the two components.

Paper:: H.W. Güsgen and J. Hertzberg: Some Fundamental Properties of Local Constraint Propagation. Artif. Intell. 36:237-247, 1988
Abstract:: In this paper, we give results about termination of local constraint propagation algorithms under possibly infinite variable coverings. The locally consistent (arc consistent) solution of a constraint problem---if such a solution exists---is shown to be unique, and a sufficient condition on propagations is given to find that solution even in the presence of arbitrary infinite variable coverings.

Paper:: W. Hammer and J. Hertzberg and H. Simon and W. Skerhut: Ein Expertensystem für Einrichtung und Endkontrolle an CNC-Maschinen. atp (Automatisierungstechnische Praxis) 30:397-400, 1988
Abstract:: Sorry, not available in electronic format!

Paper:: J. Hertzberg: Planungsmethoden der Künstlichen Intelligenz. Informatik-Spektrum 9:149-161, 1986
Abstract:: Sorry, not available in electronic format!

Invited papers, book chapters

Paper:: J. Hertzberg: Planning. in: J.G. Webster (ed.): Wiley Encyclopedia of Electrical and Electronics Engineering. Wiley, 1999, vol. 16, pp. 493-504
Abstract:: The paper comprehensively reviews the field of AI planning. (Extended Abstract, ps.gz version)
Full paper in electronic format available from the publisher. Consult the Encyclopedia of Electrical and Electronics Engineering home page.

Paper:: J. Hertzberg: Planen. In: G. Strube (ed.): Wörterbuch der Kognitionswissenschaft. Stuttgart (Klett-Cotta), 1996, pp. 501-509.
Abstract:: No abstract.

Paper:: R. Weißschnur and J. Hertzberg and H.W. Guesgen: Experiences in Solving Constraint Relaxation Networks with Boltzmann Machines. In: M. Jampel and E. Freuder and M. Maher (eds.): Over-Constrained Systems. Berlin (Springer, LNCS vol. 1106) 1996, pp. 281-292.
Abstract:: Earlier, Guesgen and Hertzberg have given a theoretical description of how to implement constraint relaxation in terms of combinatorial optimization using the concept of Boltzmann Machines. This paper sketches some lessons that an implementation of this idea has taught us about how to tailor the translation from constraint networks to Boltzmann Machines such that the resulting implementation be efficient.
Postscript Source (ps.gz, 69 KB)

Paper:: J. Hertzberg: Planning Assistance. In: P. Hoschka (ed.): Computers as Assistants. A New Generation of Support Systems. Lawrence Erlbaum 1996, pp. 255-268
Abstract:: No abstract.

Paper:: J. Hertzberg: Planen aus Sicht der Künstlichen Intelligenz: Time for a Change. In: J. Funke and A. Fritz (eds.): Neue Konzepte und Instrumente zur Planungsdiagnostik. Bonn (Deutscher Psychologen Verlag) 1995, pp. 79-96
Abstract:: Der Text skizziert Planung als ein Teilgebiet der Künstlichen Intelligenz (KI). Dieses Teilgebiet wandelt sich zur Zeit. Aus Konsequenzen dieses Wandels leitet der Text Hypothesen über Nebenbedingungen ab, denen Modelle des Planens auch außerhalb der KI Rechnung tragen müssen.

Paper:: J. Hertzberg: How to Plan under Uncertainty, and Use a Logical Action Model, too. In: H. Bunke, T. Kanade, and H. Noltemeyer (eds.): Modelling and Planning for Sensor Based Intelligent Robot Systems. Singapore (World Scientific, Machine Perception and Artificial Intelligence vol. 21) 1995, pp. 160-175
Abstract:: Planning systems are supposed to find a course of action to make given goals true in the world---and find it effectively. Logics for action and change are meant to codify intuitively correct reasoning about the effects of actions---and are unlikely to be effectively implemented. The paper shows by example a way to ground a planner in an action logic anyway: let it be limitedly correct wrt. the logic and the available run time. The logic we use allows some forms of uncertainty about the domain to be expressed, which will be shown to be in harmony with the notion of limited correctness.
Postscript Source (ps.gz, 62 KB)

Paper:: J. Hertzberg: On Building a Planning Tool Box. In: M. Ghallab and A. Milani (eds.): New Directions in in AI Planning. (Proc. EWSP'95). IOS Press, 1996, pp. 3-18
Abstract:: A planning tool box is a set of software modules implementing different planning algorithms, techniques, and representation languages. A planning system designer can use these modules---and possibly additional own ones---for building a whole family of generic or application planners. The paper explains the concept and outlines work on developing such a tool box. In particular, it identifies descriptions of planning domain characteristics as a topic in planning theory from which results are needed for building a tool box, yet are missing.
Postscript Source (ps.gz, 66 KB)

Paper:: J. Hertzberg: Was kann die Kognitionswissenschaft von der KI-Handlungsplanung lernen? KI 6:43-44, 1995
Abstract:: No abstract.

Paper:: J. Hertzberg: Anytime-Algorithmen. KI 9(1):40, 1995
Abstract:: No abstract.

Paper:: D. Bolz, T. Gordon, J. Hertzberg, and J. Schneeberger: Planen unter Unsicherheit und planbasiertes Verschönern von Grafiken. In: F. di Primio (ed.): Methoden der Künstlichen Intelligenz für Grafikanwendungen. Bonn (Addison Wesley) 1995, pp. 115-158
Abstract:: Wir fassen die TASSO-Arbeiten zum Methoden-Thema ``Planen unter Unsicherheit'' und zum Anwendungs-Thema ``Beautifizieren von Geschäftsgrafiken'' zusammen. Dabei skizzieren wir insbesondere, wie das Beautifizieren als Anwendung des Planens verstanden werden kann.

Paper:: S. Biundo, A. Günter, J. Hertzberg, J. Schneeberger, and W. Tank: Planen und Konfigurieren. In: G. Görz (ed.): Künstliche Intelligenz. Bonn (Addison Wesley) 1995, pp. 754-811. (2nd, revised print. 1st printing 1993.)
Abstract:: No abstract.

Paper:: J. Hertzberg: Theoretical Planning and its Contributions to Practical and Applied Planning (Extended Abstract). In: A.G. Cohn (ed.): 11th European Conference on Artificial Intelligence ECAI94. Wiley, 1994, pp. 811-812
Abstract:: No abstract.

Paper:: J. Hertzberg: KI-Handlungsplanung -- Woran wir arbeiten, und woran wir arbeiten sollten. In: O. Herzog, Th. Christaller, and D. Schütt (eds.): Grundlagen und Anwendungen der Künstlichen Intelligenz. 17. Fachtagung für Künstliche Intelligenz (KI'93). Springer, 1993, pp. 3-27
Abstract:: Das KI-Gebiet Handlungsplanung befindet sich in einer Phase der Neuorientierung. In Phasen von Neuorientierung ist es wichtig zu verstehen, wo man hinwollen kann. Der Aufsatz skizziert eine Perspektive ``ingenieurorientierter'' Handlungsplanung, die wesentlich basiert auf einem Raster aus Einschränkungen von Anwendungsbereichen und Anforderungen an Pläne und Planer. Er beschreibt dieses Raster, und er gibt Beispiele von Arbeiten, die sich in die Perspektive einordnen lassen.

Major conference papers

Extended versions of some of my conference papers are included in the journal or book contributions above. They are omitted here.

Paper:: A. Arghir, J. Hertzberg: Keep It Simple, Hybrid! -- A Case Study in Autonomous Office Courier Robot Control. In: M. Devy (ed.): Proc. 9th International Symposium on Intelligent Robotic Systems (SIRS 2001), 18-20 July 2001, Toulouse, France, pp. 25-33.
Abstract:: We describe the control architecture of an autonomous mobile robot, using courier tasks in office buildings as a demonstrator domain. The robot robustly performs state-of-the-art tasks such as autonomous navigation through the building, dynamically avoiding obstacles, entering rooms, passing automatic doors, and calling and riding the elevator. We give performance statistics of robot tours of about 8\,km length in our three-storied institute building. The interesting feature of the control software is its simplicity. This is achieved by combining pieces of fuzzy control, basic geometrical reasoning, target-point planning, and map-based route planning in a hybrid levelled control architecture. The control parameters proposed by low-level behaviors are fused and executed by a complex behavior, which gets selected according to the recently executed operator in a high-level mission plan. The paper sketches these pieces individually and their interplay, with emphasis on the low-level control.
Full paper in PDF

Paper:: M. Hassel, J. Hertzberg: Sensor Fusion for Localizing a Mobile Robot Outside of Buildings. In: M. Devy (ed.): Proc. 9th International Symposium on Intelligent Robotic Systems (SIRS 2001), 18-20 July 2001, Toulouse, France, pp. 477-484.
Abstract:: In this paper we show that it is possible to extend a mobile indoor service robot, making it capable of performing autonomous transportation tasks in outdoor environments. While the higher-level software (like planners, schedulers and collision avoidance) is no different than that used in indoor vehicles and therefore both well known and reliable, the sensor fusion layer is the challenging part due to the extreme ambiguity of sensor data outside most buildings. Therefore we focus on the self-localization aspect of the system, showing that it is possible to predict the robot's position with such low uncertainty that a typical transportation task in an industrial outdoor environment can be performed quickly, safely and robustly. We rely only on low cost off-the-shelf sensors (optical encoder, fiber-optical gyroscope and a laser range finder), without adjusting the robot's environment (e.g. by way of addition of artificial landmarks), making the system affordable and easy to maintain. We prove the usability of our approach through extensive tests with a real robot around our institute campus. Test drives totalling over 20km show that it is possible to find and traverse a target door in over 99% of all cases.
Full paper in PDF

Paper:: L. Paletta, S. Frintrop, J. Hertzberg: Robust localization using context in omnidirectional imaging. In: 2001 IEEE Intl. Conf. on Robotics and Automation (ICRA2001) (May 21-26, 2001, Seoul, Korea), pp. 2072-2077
Abstract:: This work presents the concept to recover and utilize the visual context in panoramic images. Omnidirectional imaging has become recently an efficient basis for robot navigation. The proposed Bayesian reasoning over local image appearances enables to reject false hypotheses which do not fit the structural constraints in corresponding feature trajectories. The methodology is proved with real image data from an office robot to dramatically increase the localization performance in the presence of severe occlusion effects, particularly in noisy environments, and to recover rotational information on the fly.
Full paper in PDF (555 kB)

Paper:: H. W. Guesgen, J. Hertzberg: Algorithms for buffering fuzzy raster maps. In: Proc. 14th Intl. FLAIRS Conference (FLAIRS-2001) (May 21-23, 2001, Key West, Florida), pp. 542-546, 2001
Abstract:: In this paper, we show how standard GIS operations like the complement, union, intersection, and buffering of maps can be made more flexible by using fuzzy set theory. In particular, we present a variety of algorithms for operations on fuzzy maps, focusing on buffer operations for fuzzy maps.
Full paper, PDF version (149 kB)

Paper:: L. Paletta, S. Frintrop, J. Hertzberg: Visual localization using Bayesian decision fusion on omnidirectional sensing. In: SPIE Conference on Sensor Fusion (16-20 April 2001, Orlando, Florida USA), pp. 58-66
Full paper in PDF (848 kB)

Paper:: H. Surmann, K. Lingemann, A. Nüchter, J. Hertzberg: A 3D laser range finder for autonomous mobile robots. In: Proc. 32nd Intl. Symp. on Robotics (ISR2001) (19-21 April 2001, Seoul, Korea), S. 153-158.
Abstract:: This paper presents a high quality, low cost 3D laser range finder designed for autonomous mobile systems. The 3D laser is built on the base of a 2D range finder by the exten-sion with a standard servo. The servo is controlled by a computer running RT-Linux. The scan resolution (approx. 5 cm) for a complete 3D scan of an area of 150 (h) x 90 (v) degree is up to 115000 points and can be grabbed in 12 seconds. Standard resolutions e.g. 150 (h) x 90 (v) degree with 22500 points are grabbed in 4 seconds. While scanning, different online algorithms for line and surface detection are applied to the data. Object segmentation and detection are done off-line after the scan. The implemented software modules detect overhanging objects blocking the path of the robot. With the proposed approach a cheap, precise, reliable and real-time capable 3D sensor for autonomous mobile robots is available and the robot navigation and recognition in real-time is improved.
Full paper, PDF version (2.9 MB)

Paper:: J. Hertzberg and F. Schönherr: Concurrency in the DD&P Robot Control Architecture. In: Proc. Int. NAISO Cong. Information Science Innovations (ISI'2001), March 17-21, 2001, Dubai, pp. 1079-1085
Abstract:: DD&P is a hybrid robot control architecture currently under development. It features logically and physically concurrent robot control components, the role of action plans as media of high-level advice and communication rather than rigid control regimes, and a direct structural coupling of the high-level planning and the low-level dynamic execution component, making DD&P a two-level architecture. The paper focuses on DD&P's concurrency feature: we design and implement behaviors as well as deliberative control components in DD&P as concurrent actors. Logical concurrency suggests to design extremely modular DD&P robot control programs; physical concurrency allows multiple onboard processors and radio-linked off-board computers to be exploited. The paper contributes a method for designing and implementing concurrent robot control systems in a modern hybrid architecture.
Full paper in PDF

Paper:: H. Streich, A. Schmitz, J. Hertzberg, and F. Kirchner: Implementierung der höheren Steuerung einer mehrsegmentigen autonomen Kanalroboterplattform. Robotik 2000: Leistungsstand -- Anwendungen -- Visionen --Trends. Tagung Berlin, 29./30. Juni 2000. Düsseldorf (VDI-Verlag) 2000, pp.209-214.
Abstract:: Wir beschreiben die Struktur der höheren Steuerungssoftware für eine kinematisch komplexe, mehrsegmentige, mobile Roboterplattform, deren Aufgabe es ist, durch Abwasserkanäle autonom zu navigieren. Durch die Strukturierung der Steuerung in Kontexte, Aktivitäten und Tasks ist es möglich, auf oberer Ebene eine angemessen abstrakte Schnittstelle zur Missionssteuerung anzubieten, auf der unteren Ebene aber Kontroll-Bausteine zu verwenden, die ausreichend feinkörnig sind, dass für sie Realzeitanforderungen spezifiziert und durchgesetzt werden können.
Full paper in PDF

Paper:: F. Schönherr, J. Hertzberg, and W. Burgard: Probabilistic Mapping of Unexpected Objects by a Mobile Robot. In: Proc. IEEE/RSJ Int. Conf. Intell. Robots and Systems (IROS'99), IEEE Press 1999, vol.1, pp. 474-481.
Abstract:: Map learning methods are generally designed to learn from scratch and start with zero knowledge about the state of the world. In this paper, we present a technique for extending a given metric map of the environment by objects of a known type, where localization and perception of the robot is allowed to be uncertain. The advantage of our approach is that it allows the robot to estimate its own position in the given outline of the environment and thus to estimate the position of the objects not contained in the map. The method relies on partially observable Markov decision processes as well as on the Baum-Welch algorithm. It has been implemented and evaluated in several simulation experiments and also in a real-world sewage pipe network. The experimental results demonstrate that our approach can efficiently and accurately estimate the position of unexpected objects. Because of the probabilistic nature of the underlying techniques, our method can deal with noisy sensors as well as with large odometry errors which generally occur when deploying a robot in a sewerage pipe system.
Full paper currently not available in electronic form.

Paper:: J. Hertzberg and F. Schönherr: On Correcting Sewer Robots' Errors by Reasoning. in: Proc. 12th Intl. Florida AI Research Soc. Conf. (FLAIRS-99) (A.N. Kumar and I. Russell, eds.), AAAI Press 1999, pp. 417-421.
Abstract:: When inspecting sewers, it is required that discovered objects like damages or previously unknown laterals are entered into the existing sewer map at the correct metric position. On the other hand, odometry conditions in sewers can be very poor, and they may vary greatly with slip, slope, and sewage current. The paper describes a simple reasoning method for correcting the odometry error according to local conditions. The method is based on correcting past measurements, and hence the corresponding new map entries, according to safe localization points at individual landmarks and to a model of the local odometry error or errors in between the localization points. A sketch of the intended application of the method in an autonomous, multi-segment, articulated sewer robot is given.
Full paper in PDF

Paper:: J. Hertzberg, F. Kirchner, U. Licht, E. Rome F. Schönherr, and T. Christaller: MAKRO -- Bau einer mehrsegmentigen autonomen Kanalroboterplattform. in: Pipeline Integrity -- Uneingeschränkte Verfügbarkeit und unbegrenzte Nutzungsdauer von Rohrleitungen und Rohrnetzen? (J. Lenz, ed.) Vulkan-Verlag, 1999, pp.504-513
Abstract:: No abstract
Full paper in PDF

Paper:: J. Hertzberg, H. Jaeger, Ph. Morignot, U.R. Zimmer: A Framework for Plan Execution in Behavior-Based Robots. in: Proc. of the 1998 IEEE Int. Symp. on Intell. Control (ISIC-98); Gaithersburg, MD, Sept. 1998, pp.8-13
Abstract:: This paper presents a conceptual architecture for autonomous robots that integrates behavior-based and goal-directed action as by following a traditional action plan. Dual Dynamics is the formalism for describing behavior-based action. Partial-order propositional plans, which get generated by GRAPHPLAN, are used as a basis for acting goal-directedly; the concept is suitable for using other planning methods and plan representations, though. The paper presents the corresponding action and plan representations at the plan side and at the behavior side. Moreover, it describes how behavior-based action is biased towards executing a plan and how information from the behavior side is fed back to the plan side to determine progress in the plan execution.
Full Paper in PDF

Paper:: J. Hertzberg, Th. Christaller, F. Kirchner, U. Licht and E. Rome: Sewer Robotics. in: From Animals to Animats 5 -- Proc. 5th Intl. Conf. on Simulation of Adaptive Behavior (SAB-98) (R. Pfeifer, B. Blumberg, J.-A. Meyer, and S.W. Wilson, eds.), MIT Press 1998, pp.427-436
Abstract:: We expose the application area of building and controlling autonomous service robots for performing inspection tasks in sewerage systems that are inaccessible for humans. The problem is explained, the physical and legal boundary conditions for operating sewer robots are described, the state of current sewer inspection technology is sketched, and the need for using advanced technology in this area motivated. Based on that, we identify AI research and technology topics that are crucial for sewer robotics and give pointers to related work, including work that we have been doing since the last four years.
Full paper in PDF

Paper:: J. Hertzberg, Th. Christaller, F. Kirchner, U. Licht and E. Rome: MAKRO -- Serviceroboter im Kanalbetrieb. in: J. Lenz (ed.): Globalisierung der Märkte und internationale Arbeitsteilung - auch im Rohrleitungsbau? Vulkan-Verlag, 1998, pp. 643-652
Abstract:: No abstract.
Full paper in PDF

Paper: J. Hertzberg and F. Kirchner: Landmark-Based Autonomous Navigation in Sewerage Pipes. in: Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT `96), IEEE Press, 1996, pp. 68-73.
Abstract:: We describe a method for an autonomous mobile robot to navigate through a system of sewerage pipes. Landmarks signalling positions in the pipe system have to be detected and classified, where classification is allowed to be unreliable. Self localization is interpreted as a partially observable Markov decision problem and solved accordingly. The method is implemented and used on a prototype robot platform operating in a dry sewage pipe test network.
Postscript Source (ps.gz, 1150 KB)

Paper:: R. Weißschnur, J. Hertzberg, and H.W. Guesgen: In Search of a Perfect Boltzmann Machine for CSPs. In: Proc. FLAIRS-96, Key West, Florida, 1996, pp. 35-39.
Abstract:: Earlier, Guesgen and Hertzberg have given a theoretical description of how to implement constraint relaxation in terms of combinatorial optimization using the concept of Boltzmann Machines (BMs). This paper sketches some lessons that an implementation of this idea has taught us about how to tailor the translation from constraint networks to BMs such that the resulting implementation be efficient.

Paper:: S. Thiébaux and J. Hertzberg: A Semi-Reactive Planner Based on a Possible Models Action Formalization. In: J. Hendler (ed.): Artificial Intelligence Planning Systems: Proceedings of the First International Conference (AIPS92). Morgan Kaufmann, 1992, pp. 228-235
Abstract:: We describe the basics of a semi-reactive planner that is based on a possible models action formalization. By semi-reactive planner, we mean a planning system that is able to incrementally generate plans in a classical way, thereby extending, when necessary, the compiled knowledge available for reactive execution. The action formalization, which is a correction of the possible worlds approach by Ginsberg and Smith, allows one to cope with lacking information, ambiguity and context dependency in the planning domain by considering different courses of action. This enables the planner to handle certain forms of uncertainty at planning time. Plans developed under this model are compiled into finite state automata for execution, as a basis for reacting to less expected effects at execution time. Moreover, we explain how replanning fits into the framework, thereby opening it towards handling unexpected effects which occur during plan execution, and for which no reaction plan is available.

Paper:: J. Hertzberg and H.W. Güsgen: Transforming Constraint Relaxation Networks into Boltzmann Machines. In: Th. Christaller (ed.): Proc. GWAI-91. Berlin (Springer, Informatik Fachberichte 285) 1991, pp. 244-253.
Abstract:: We describe how to transform constraint networks---which may involve a particular form of constraint relaxation---into corresponding Boltzmann machines, thereby viewing constraint satisfaction as a problem of combinatorial optimization. We discuss feasibility and order preservingness of the consensus function used and give a necessary and sufficient condition for a locally optimal configuration to correspond to a solution of the constraint network.

Paper:: S. Materne and J. Hertzberg: MTMM - Correcting and Extending Time Map Management. In: J. Hertzberg (ed.): European Workshop on Planning. EWSP '91, Sankt Augustin, FRG, March 1991, Proceedings. Berlin (Springer, LNAI vol. 522) 1991, pp. 88-99.
Abstract:: We present the concept of the deductive temporal data base MTMM, that corrects and extends Dean's [1985] TMM in a number of aspects, allowing complete and consistent temporal inferences from a temporal model that may be given in a quantitative or qualitative way. The user is allowed to incorporate a certain form of inconsistency in the time map; we will demonstrate that this is a useful feature. Moreover, the user may withdraw information at any time. The expressiveness of the temporal logic of the MTMM is equivalent to convex subsets of Allen's relations. We present the basics of the MTMM, going into the concepts of persistence, persistence clipping, and validity in a bit more detail.

Paper:: J. Hertzberg and A. Horz: Towards a Theory of Conflict Detection and Resolution in Nonlinear Plans. In: Proc. IJCAI-89, 1989, pp. 937-942.
Abstract:: This paper deals with a well known problem in AI planning: detecting and resolving conflicts in nonlinear plans. We sketch a theory of restricted conflict handling in classical nonlinear planners. By relaxing the restrictions, we develop a more general concept of conflicts suggesting practical and theoretical limitations of conflict handling in nonlinear plans.

Paper:: J. Hertzberg, H.W. Güsgen, A. Voß, M. Fidelak and H. Voß: Relaxing Constraint Networks to Resolve Inconsistencies. In: W. Hoeppner (ed.): Proc. GWAI-88. Berlin (Springer, Informatik Fachberichte 181) 1988, pp. 61-65.
Abstract:: This paper deals with the following problem: What is to do if a constraint problem is inconsistent but one wants to solve it anyway? Concentrating on locally consistent solutions, it answeres the questions: What is a relaxation of a constraint problem? Do solutions of relaxed problems lead to a solution of the original one? Does a locally consistent relaxation always exist? How to obtain such a relaxation efficiently? As an answer to the last question, we present an algorithm which has been implemented as an extension of the CONSAT constraint system.