ThisvolumecontainstheresearchpaperspresentedatKI2005,the28thGerman Conference on Arti?cial Intelligence, held September 11 14, 2005 in Koblenz, Germany. KI 2005 was part of the International Conference Summer Koblenz 2005, which included conferences covering a broad spectrum of topics that are all related to AI: tableau-based reasoningmethods (TABLEAUX), multi-agent s- tems (MATES), automated reasoningand knowledgerepresentation(FTP), and software engineering and formal methods (SEFM). The Program Committee received 113 submissions from 22 countries. Each paperwasreviewedbythreereferees;afteranintensivediscussionabouttheb- derline papers during the online meeting of the Program Committee, 29 papers were accepted for publication in this proceedings volume. The program included three outstanding keynote talks: Ian Horrocks (U- versityofManchester,UK),LucSteels(UniversityofBrusselsandSony)and- bastian Thrun (Stanford University), who covered topics like logical foundation, cognitive abilities of multi-agent systems and the DARPA Grand Challenge. KI 2005 also included two excellent tutorials: Techniques in Evolutionary Robotics and Neurodynamics (Frank Pasemann, Martin Hulse, Ste?en Wis- mann and Keyan Zahedi) and Connectionist Knowledge Representation and Reasoning (Barbara Hammer and Pascal Hitzler). Many thanks to the tu- rial presenters and the tutorial chair Joachim Hertzberg. Peter Baumgartner, in hisroleasaworkshopchair,collected11workshopsfromallareasofAIresearch, whichalsoincludes ameeting ofthe GermanPriorityProgramonKooperierende Teams mobiler Roboter in dynamischen Umgebungen. I want to sincerely thank all the authors who submitted their work for c- sideration and the Program Committee members and the additional referees for theirgreate?ortandprofessionalworkinthereviewandselectionprocess.Their names are listed on the following pages. Invited Talks.- Hierarchy in Fluid Construction Grammars.- Description Logics in Ontology Applications.- 175 Miles Through the Desert.- Knowledge Representation and Reasoning.- A New n-Ary Existential Quantifier in Description Logics.- Subsumption in w.r.t. Hybrid TBoxes.- Dependency Calculus: Reasoning in a General Point Relation Algebra.- Temporalizing Spatial Calculi: On Generalized Neighborhood Graphs.- Machine Learning.- Design of Geologic Structure Models with Case Based Reasoning.- Applying Constrained Linear Regression Models to Predict Interval-Valued Data.- On Utilizing Stochastic Learning Weak Estimators for Training and Classification of Patterns with Non-stationary Distributions.- Noise Robustness by Using Inverse Mutations.- Diagnosis.- Development of Flexible and Adaptable Fault Detection and Diagnosis Algorithm for Induction Motors Based on Self-organization of Feature Extraction.- Computing the Optimal Action Sequence by Niche Genetic Algorithm.- Diagnosis of Plan Execution and the Executing Agent.- Automatic Abstraction of Time-Varying System Models for Model Based Diagnosis.- Neural Networks.- Neuro-Fuzzy Kolmogorov's Network for Time Series Prediction and Pattern Classification.- Design of Optimal Power Distribution Networks Using Multiobjective Genetic Algorithm.- New Stability Results for Delayed Neural Networks.- Planning.- Metaheuristics for Late Work Minimization in Two-Machine Flow Shop with Common Due Date.- An Optimal Algorithm for Disassembly Scheduling with Assembly Product Structure.- Hybrid Planning Using Flexible Strategies.- Controlled Reachability Analysis in AI Planning: Theory and Practice.- Robotics.- Distributed Multi-robot Localization Based on Mutual Path Detection.- Modeling Moving Objects in a Dynamically Changing RobotApplication.- Heuristic-Based Laser Scan Matching for Outdoor 6D SLAM.- A Probabilistic Multimodal Sensor Aggregation Scheme Applied for a Mobile Robot.- Behavior Recognition and Opponent Modeling for Adaptive Table Soccer Playing.- Cognitive Modelling / Philosopy / Natural Language.- Selecting What Is Important: Training Visual Attention.- Self-sustained Thought Processes in a Dense Associative Network.- Why Is the Lucas-Penrose Argument Invalid?.- On the Road to High-Quality POS-Tagging. Ulrich Furbach ist Professor für Künstliche Intelligenz an der Universität Koblenz-Landau. Seine Forschungsgebiete umfassen Automatisches Schließen, Agenten und Robotik sowie Frage-Antwort-Systeme. Er ist an der TU München habilitiert, an der Uni der Bundeswehr promoviert und ist Gründer und Gesellschafter des KI-Unternehmens wizAI.
