\n \n \n
\n
\n\n \n \n \n \n \n \n Planning and Verification in the Agent Language Golog.\n \n \n \n \n\n\n \n Jens Claßen.\n\n\n \n\n\n\n Ph.D. Thesis, Department of Computer Science, RWTH Aachen University, 2013.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@PHDTHESIS{Cla:PHD2013,\n author = {Jens Cla{\\ss}en},\n title = {Planning and Verification in the Agent Language\n {G}olog},\n school = {Department of Computer Science, RWTH Aachen\n University},\n year = {2013},\n abstract = {The action programming language Golog has proven to\n be a useful means for the high-level control of\n autonomous agents such as mobile robots. It is based\n on the Situation Calculus, a dialect of classical\n first-order logic, that is used to encode dynamic\n domains through logical axioms. Perhaps the greatest\n advantage of Golog is that a user can write programs\n which constrain the search for an executable plan in\n a flexible manner. However, when general planning is\n needed, Golog supports this only in principle, but\n does not measure up with state-of-the-art planners,\n most of which are based on the plan language\n PDDL. On the other hand, planning formalisms and\n systems lack the expressiveness of Golog that make\n it suited for realistic scenarios of agents with\n partial world knowledge acting in dynamic\n environments. We therefore propose an integration of\n Golog and planning where planning subtasks\n encountered during the execution of a Golog program\n are referred to a PDDL planner, thus combining\n Golog{\\textquoteright}s expressiveness with the\n efficiency of modern planners. The theoretical\n justification for such an embedding is provided in\n the form of relating state updates in PDDL to the\n progression of a certain form of theories of the\n modal Situation Calculus variant ES. We complement\n these results with an empirical evaluation that\n shows that equipping Golog with a PDDL planner\n indeed pays off in terms of the runtime\n performance. Moreover, before deploying a Golog\n program onto a robot, it is often desirable to\n verify that certain requirements are met, typical\n examples including safety, liveness and fairness\n conditions. Since autonomous robots typically\n perform open-ended tasks, the corresponding control\n programs are often non-terminating. Analyzing such\n programs so far requires manual, meta-theoretic\n arguments involving complex fixpoint constructions,\n which is tedious and error-prone. In this thesis, we\n propose an extension to ES that includes new modal\n operators to express temporal properties of Golog\n programs. We then provide algorithms for the\n automated verification of such properties, relying\n on a newly introduced graph representation for Golog\n programs which enables a systematic exploration of\n the state space. Similar to other forms of reasoning\n in the Situation Calculus, our verification methods\n ultimately reduce to classical first-order theorem\n proving.},\n url = {http://darwin.bth.rwth-aachen.de/opus3/volltexte/2013/4809/}\n}\n\n
\n
\n\n\n
\n The action programming language Golog has proven to be a useful means for the high-level control of autonomous agents such as mobile robots. It is based on the Situation Calculus, a dialect of classical first-order logic, that is used to encode dynamic domains through logical axioms. Perhaps the greatest advantage of Golog is that a user can write programs which constrain the search for an executable plan in a flexible manner. However, when general planning is needed, Golog supports this only in principle, but does not measure up with state-of-the-art planners, most of which are based on the plan language PDDL. On the other hand, planning formalisms and systems lack the expressiveness of Golog that make it suited for realistic scenarios of agents with partial world knowledge acting in dynamic environments. We therefore propose an integration of Golog and planning where planning subtasks encountered during the execution of a Golog program are referred to a PDDL planner, thus combining Golog\\textquoterights expressiveness with the efficiency of modern planners. The theoretical justification for such an embedding is provided in the form of relating state updates in PDDL to the progression of a certain form of theories of the modal Situation Calculus variant ES. We complement these results with an empirical evaluation that shows that equipping Golog with a PDDL planner indeed pays off in terms of the runtime performance. Moreover, before deploying a Golog program onto a robot, it is often desirable to verify that certain requirements are met, typical examples including safety, liveness and fairness conditions. Since autonomous robots typically perform open-ended tasks, the corresponding control programs are often non-terminating. Analyzing such programs so far requires manual, meta-theoretic arguments involving complex fixpoint constructions, which is tedious and error-prone. In this thesis, we propose an extension to ES that includes new modal operators to express temporal properties of Golog programs. We then provide algorithms for the automated verification of such properties, relying on a newly introduced graph representation for Golog programs which enables a systematic exploration of the state space. Similar to other forms of reasoning in the Situation Calculus, our verification methods ultimately reduce to classical first-order theorem proving.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n On Decidable Verification of Non-terminating Golog Programs.\n \n \n \n \n\n\n \n Jens Claßen; Martin Liebenberg; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In Jianmin Ji; Hannes Strass; and Xun Wang., editor(s),
Proceedings of the 10th International Workshop on Nonmonotonic Reasoning, Action and Change (NRAC 2013), pages 13–20, 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@INPROCEEDINGS{ClaLieLak:NRAC2013,\n title = {On Decidable Verification of Non-terminating {G}olog\n Programs},\n author = {Jens Cla{\\ss}en and Martin Liebenberg and Gerhard\n Lakemeyer},\n booktitle = {Proceedings of the 10th International Workshop on\n Nonmonotonic Reasoning, Action and Change (NRAC\n 2013)},\n year = {2013},\n editor = {Jianmin Ji and Hannes Strass and Xun Wang},\n pages = {13--20},\n abstract = {The high-level action programming language Golog has\n proven to be a useful means for the control of\n autonomous agents such as mobile robots. Usually,\n such agents perform open-ended tasks, and their\n control programs are hence non-terminating. Before\n deploying such a program to the robot, it is often\n desirable if not crucial to verify that it meets\n certain requirements, preferably by means of an\n automated method. For this purpose, Cla{\\ss}en and\n Lakemeyer recently introduced algorithms for the\n verification of temporal properties of\n non-terminating Golog programs, based on the\n first-order modal Situation Calculus variant \\ES,\n and regression-based reasoning. However, while\n Golog{\\textquoteright}s high expressiveness is a\n desirable feature, it also means that their\n verification procedures cannot be guaranteed to\n terminate in general. In this paper, we address this\n problem by showing that, for a relevant subset, the\n verification of non-terminating Golog programs is\n indeed decidable, which is achieved by means of\n three restrictions. First, we use the ES variant of\n a decidable two-variable fragment of the Situation\n Calculus that was introduced by Gu and\n Soutchanski. Second, we have to restrict the Golog\n program to contain ground action only. Finally, we\n consider special classes of successor state axioms,\n namely the context-free ones and those that only\n admit local effects.},\n url = {https://kbsg.rwth-aachen.de/~classen/pub/ClassenLiebenbergLakemeyer2013.pdf}\n}\n\n
\n
\n\n\n
\n The high-level action programming language Golog has proven to be a useful means for the control of autonomous agents such as mobile robots. Usually, such agents perform open-ended tasks, and their control programs are hence non-terminating. Before deploying such a program to the robot, it is often desirable if not crucial to verify that it meets certain requirements, preferably by means of an automated method. For this purpose, Claßen and Lakemeyer recently introduced algorithms for the verification of temporal properties of non-terminating Golog programs, based on the first-order modal Situation Calculus variant \\ES, and regression-based reasoning. However, while Golog\\textquoterights high expressiveness is a desirable feature, it also means that their verification procedures cannot be guaranteed to terminate in general. In this paper, we address this problem by showing that, for a relevant subset, the verification of non-terminating Golog programs is indeed decidable, which is achieved by means of three restrictions. First, we use the ES variant of a decidable two-variable fragment of the Situation Calculus that was introduced by Gu and Soutchanski. Second, we have to restrict the Golog program to contain ground action only. Finally, we consider special classes of successor state axioms, namely the context-free ones and those that only admit local effects.\n
\n\n\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Simulation-based approach for avoiding external faults.\n \n \n \n \n\n\n \n Naveed Akhtar; Anastassia Küstenmacher; Paul G. Plöger; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In 11/2013 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings {KBSG_300,\n title = {Simulation-based approach for avoiding external faults},\n journal = {16th International Conference on Advanced Robotics, ICAR'13},\n year = {2013},\n month = {11/2013},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/ExternalFaultsSimBasedAppr.pdf},\n author = {Naveed Akhtar and Anastassia K{\\"u}stenmacher and Paul G. Pl{\\"o}ger and Lakemeyer, Gerhard}\n}\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Improving Robustness of Task Execution Against External Faults Using Simulation Based Approach.\n \n \n \n \n\n\n \n Anastassia Küstenmacher; Paul G. Plöger; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In 10/2013 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings {KBSG_299,\n title = {Improving Robustness of Task Execution Against External Faults Using Simulation Based Approach},\n journal = {24nd International Workshop on Principles of Diagnosis DX'13},\n year = {2013},\n month = {10/2013},\n abstract = {Robots interacting in complex and cluttered\n environments may face unexpected situations referred\n to as external faults which prohibit the successful\n completion of their tasks. In order to function in a\n more robust manner, robots need to recognise these\n faults and learn how to deal with them in the\n future. We present a simulation-based technique to\n avoid external faults occurring during execusion\n releasing actions of a robot. Our technique utilizes\n simulation to generate a set of labeled examples\n which are used by a histogram algorithm to compute a\n safe region. A safe region consists of a set of\n releasing states of an object that correspond to\n successful performances of the action. This\n technique also suggests a general solution to avoid\n the occurrence of external faults for not only the\n current, observable object but also for any other\n object of the same shape but different size.},\n keywords = {faults in robotics, simulation},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/kuestenmacher_DX13_0.pdf},\n author = {Anastassia K{\\"u}stenmacher and Paul G. Pl{\\"o}ger and Lakemeyer, Gerhard},\n}\n
\n
\n\n\n
\n Robots interacting in complex and cluttered environments may face unexpected situations referred to as external faults which prohibit the successful completion of their tasks. In order to function in a more robust manner, robots need to recognise these faults and learn how to deal with them in the future. We present a simulation-based technique to avoid external faults occurring during execusion releasing actions of a robot. Our technique utilizes simulation to generate a set of labeled examples which are used by a histogram algorithm to compute a safe region. A safe region consists of a set of releasing states of an object that correspond to successful performances of the action. This technique also suggests a general solution to avoid the occurrence of external faults for not only the current, observable object but also for any other object of the same shape but different size.\n
\n\n\n
\n\n\n \n\n\n
\n
\n\n \n \n \n \n \n Unexpected Situations in Service Robot Environment: Classification and Reasoning Using Naive Physics.\n \n \n \n\n\n \n Anastassia Küstenmacher; Naveed Akhtar; Paul G. Plöger; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In Eindhoven, Netherlands, 07/2013 2013. \n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings {KBSG_298,\n title = {Unexpected Situations in Service Robot Environment: Classification and Reasoning Using Naive Physics},\n journal = {17th annual RoboCup International Symposium},\n year = {2013},\n month = {07/2013},\n address = {Eindhoven, Netherlands},\n abstract = {Despite perfect functioning of its internal components, a robot can\r\nbe unsuccessful in performing its tasks because of unforeseen situations. Mostly\r\nthese situations arise from the interaction of a robot with its ever-changing environment.\r\nIn this paper we refer to these unsuccessful operations as external\r\nunknown faults. We reason along the most frequent failures in typical scenarios\r\nwhich we observed during real-world demonstrations and competitions using our\r\nCare-O-bot III robot. These events take place in an apartment-like environment.\r\nWe create four different - for now adhoc - fault classes, which refer to faults\r\ncaused by a) disturbances, b) imperfect perception, c) inadequate planning or d)\r\nchaining of action sequences. These four fault classes can then be mapped to a\r\nhandful of partly known, partly extended fault handling techniques.\r\nIn addition to existing techniques we propose an approach that uses naive physics\r\nconcepts to find information about these kinds of situations. Here the naive physics\r\nknowledge is represented by the physical properties of objects which are formalized\r\nin a logical framework. The proposed approach applies a qualitative version\r\nof physical laws to these properties to reason about the fault. By interpreting the\r\nresults the robot finds the information about the situations which can cause the\r\nfault.We apply this approach to scenarios in which a robot performs manipulation\r\ntasks (pick and place). The results show that naive physics hold great promises\r\nfor reasoning about unknown external faults in the field of robotics.},\n keywords = {faults in robotics, naive physics, unexpected situations},\n author = {Anastassia K{\\"u}stenmacher and Naveed Akhtar and Paul G. Pl{\\"o}ger and Lakemeyer, Gerhard}\n}\n
\n
\n\n\n
\n Despite perfect functioning of its internal components, a robot can be unsuccessful in performing its tasks because of unforeseen situations. Mostly these situations arise from the interaction of a robot with its ever-changing environment. In this paper we refer to these unsuccessful operations as external unknown faults. We reason along the most frequent failures in typical scenarios which we observed during real-world demonstrations and competitions using our Care-O-bot III robot. These events take place in an apartment-like environment. We create four different - for now adhoc - fault classes, which refer to faults caused by a) disturbances, b) imperfect perception, c) inadequate planning or d) chaining of action sequences. These four fault classes can then be mapped to a handful of partly known, partly extended fault handling techniques. In addition to existing techniques we propose an approach that uses naive physics concepts to find information about these kinds of situations. Here the naive physics knowledge is represented by the physical properties of objects which are formalized in a logical framework. The proposed approach applies a qualitative version of physical laws to these properties to reason about the fault. By interpreting the results the robot finds the information about the situations which can cause the fault.We apply this approach to scenarios in which a robot performs manipulation tasks (pick and place). The results show that naive physics hold great promises for reasoning about unknown external faults in the field of robotics.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Aspects of Integrating Diverse Software into Robotic Systems.\n \n \n \n \n\n\n \n Tim Niemueller; Gerhard Lakemeyer; and Alexander Ferrein.\n\n\n \n\n\n\n In
ICRA 2013 - 8th Workshop on Software Development and Integration in Robotics (SDIR), Karlsruhe, Germany, 05/2013 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings {KBSG_292,\n title = {Aspects of Integrating Diverse Software into Robotic Systems},\n booktitle = {ICRA 2013 - 8th Workshop on Software Development and Integration in Robotics (SDIR)},\n year = {2013},\n month = {05/2013},\n address = {Karlsruhe, Germany},\n url = {http://www.fawkesrobotics.org/publications/2013/diverse-sw-integration-icra2013-sdir/},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/diverse-sw-integration-icra2013-sdir.pdf},\n author = {Tim Niemueller and Lakemeyer, Gerhard and Ferrein, Alexander}\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Life-long Learning Perception using Cloud Database Technology.\n \n \n \n \n\n\n \n Tim Niemueller; Stefan Schiffer; Gerhard Lakemeyer; and Safoura Rezapour-Lakani.\n\n\n \n\n\n\n In
IROS 2013 - Cloud Robotics Workshop, Tokyo, Japan, 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings {longterm-perception-db-iros2013,\n title = {Life-long Learning Perception using Cloud Database Technology},\n booktitle = {IROS 2013 - Cloud Robotics Workshop},\n year = {2013},\n address = {Tokyo, Japan},\n abstract = {Autonomous mobile robots in household environments have to cope with many different kinds of objects which they must detect, recognize, and manipulate. Over their lifetime, the robots must adapt to new objects and incorporate new perception methods. In this paper we present a system for life-long learning of training data and perception method parameters using a document-oriented, schema-less database technology that is typically used in cloud computing applications. Not only can a single robot extend and increase its data volume continuously over time, but it can also potentially share this very dataset with multiple other robots through the cloud.},\n keywords = {Cloud Robotics, Robot Database, Robotic Perception},\n url = {http://www.roboearth.org/iros2013},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/longterm-perception-db-iros2013.pdf},\n author = {Tim Niemueller and Schiffer, Stefan and Lakemeyer, Gerhard and Safoura Rezapour-Lakani}\n}\n
\n
\n\n\n
\n Autonomous mobile robots in household environments have to cope with many different kinds of objects which they must detect, recognize, and manipulate. Over their lifetime, the robots must adapt to new objects and incorporate new perception methods. In this paper we present a system for life-long learning of training data and perception method parameters using a document-oriented, schema-less database technology that is typically used in cloud computing applications. Not only can a single robot extend and increase its data volume continuously over time, but it can also potentially share this very dataset with multiple other robots through the cloud.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Proposal for Advancements to the LLSF in 2014 and beyond.\n \n \n \n \n\n\n \n Tim Niemueller; Gerhard Lakemeyer; Alexander Ferrein; Sebastian Reuter; Daniel Ewert; Sabina Jeschke; Dirk Pensky; and Ulrich Karras.\n\n\n \n\n\n\n In
ICAR 2013 - 1st Workshop on Developments in RoboCup Leagues, Montevideo, Uruguay, 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings {KBSG_304,\n title = {Proposal for Advancements to the LLSF in 2014 and beyond},\n booktitle = {ICAR 2013 - 1st Workshop on Developments in RoboCup Leagues},\n year = {2013},\n address = {Montevideo, Uruguay},\n url = {https://www.carologistics.org/publications/2013/llsf2014-wdrl2013/},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/llsf2014-wdrl2013.pdf},\n author = {Tim Niemueller and Lakemeyer, Gerhard and Ferrein, Alexander and Sebastian Reuter and Daniel Ewert and Sabina Jeschke and Dirk Pensky and Ulrich Karras}\n}\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Towards Deliberative Active Perception using Persistent Memory.\n \n \n \n \n\n\n \n Tim Niemueller; Nichola Abdo; Andreas Hertle; Gerhard Lakemeyer; Wolfram Burgard; and Bernhard Nebel.\n\n\n \n\n\n\n In
IROS 2013 - Workshop on AI-based Robotics, Tokyo, Japan, 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings {KBSG_303,\n title = {Towards Deliberative Active Perception using Persistent Memory},\n booktitle = {IROS 2013 - Workshop on AI-based Robotics},\n year = {2013},\n address = {Tokyo, Japan},\n abstract = {Task coordination for autonomous mobile service robots typically involves a substantial amount of background knowledge and explicit action sequences to acquire the relevant information nowadays. We strive for a system which, given a task, is capable of reasoning about task-relevant knowledge to automatically determine whether that knowledge is sufficient. If missing or uncertain, the robot shall decide autonomously on the actions to gain or improve that knowledge. In this paper we present our baseline system implementing the foundations for these capabilities. The robot has to analyze a tabletop scene and increase its object type confidence. It plans motions to observe the scene from multiple perspectives, combines the acquired data, and performs a recognition step on the merged input.},\n keywords = {Active Perception, Hybrid Reasoning, Robot Database, Robotic Perception, Task Planning},\n url = {http://robohow.eu/workshops/ai-based-robotics-iros-2013},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/hybris-c1-baseline-iros2013.pdf},\n author = {Tim Niemueller and Nichola Abdo and Andreas Hertle and Lakemeyer, Gerhard and Wolfram Burgard and Nebel, Bernhard}\n}\n\n\n
\n
\n\n\n
\n Task coordination for autonomous mobile service robots typically involves a substantial amount of background knowledge and explicit action sequences to acquire the relevant information nowadays. We strive for a system which, given a task, is capable of reasoning about task-relevant knowledge to automatically determine whether that knowledge is sufficient. If missing or uncertain, the robot shall decide autonomously on the actions to gain or improve that knowledge. In this paper we present our baseline system implementing the foundations for these capabilities. The robot has to analyze a tabletop scene and increase its object type confidence. It plans motions to observe the scene from multiple perspectives, combines the acquired data, and performs a recognition step on the merged input.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n On Decidable Verification of Non-terminating Golog Programs.\n \n \n \n \n\n\n \n Jens Claßen; Martin Liebenberg; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In Jianmin Ji; Hannes Strass; and Xun Wang., editor(s),
Proceedings of the 10th International Workshop on Nonmonotonic Reasoning, Action and Change (NRAC 2013), pages 13–20, Beijing, China, 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@INPROCEEDINGS{LiebenbergNRAC13,\n title = {On Decidable Verification of Non-terminating {G}olog\n Programs},\n author = {Jens Cla{\\ss}en and Martin Liebenberg and Gerhard\n Lakemeyer},\n booktitle = {Proceedings of the 10th International Workshop on\n Nonmonotonic Reasoning, Action and Change (NRAC\n 2013)},\n year = {2013},\n address = {Beijing, China},\n editor = {Jianmin Ji and Hannes Strass and Xun Wang},\n pages = {13--20},\n abstract = {The high-level action programming language Golog has\n proven to be a useful means for the control of\n autonomous agents such as mobile robots. Usually,\n such agents perform open-ended tasks, and their\n control programs are hence non-terminating. Before\n deploying such a program to the robot, it is often\n desirable if not crucial to verify that it meets\n certain requirements, preferably by means of an\n automated method. For this purpose, Cla{\\ss}en and\n Lakemeyer recently introduced algorithms for the\n verification of temporal properties of\n non-terminating Golog programs, based on the\n first-order modal Situation Calculus variant \\ES,\n and regression-based reasoning. However, while\n Golog{\\textquoteright}s high expressiveness is a\n desirable feature, it also means that their\n verification procedures cannot be guaranteed to\n terminate in general. In this paper, we address this\n problem by showing that, for a relevant subset, the\n verification of non-terminating Golog programs is\n indeed decidable, which is achieved by means of\n three restrictions. First, we use the ES variant of\n a decidable two-variable fragment of the Situation\n Calculus that was introduced by Gu and\n Soutchanski. Second, we have to restrict the Golog\n program to contain ground action only. Finally, we\n consider special classes of successor state axioms,\n namely the context-free ones and those that only\n admit local effects.},\n url = {https://kbsg.rwth-aachen.de/~classen/pub/ClassenLiebenbergLakemeyer2013.pdf}\n}\n\n
\n
\n\n\n
\n The high-level action programming language Golog has proven to be a useful means for the control of autonomous agents such as mobile robots. Usually, such agents perform open-ended tasks, and their control programs are hence non-terminating. Before deploying such a program to the robot, it is often desirable if not crucial to verify that it meets certain requirements, preferably by means of an automated method. For this purpose, Claßen and Lakemeyer recently introduced algorithms for the verification of temporal properties of non-terminating Golog programs, based on the first-order modal Situation Calculus variant \\ES, and regression-based reasoning. However, while Golog\\textquoterights high expressiveness is a desirable feature, it also means that their verification procedures cannot be guaranteed to terminate in general. In this paper, we address this problem by showing that, for a relevant subset, the verification of non-terminating Golog programs is indeed decidable, which is achieved by means of three restrictions. First, we use the ES variant of a decidable two-variable fragment of the Situation Calculus that was introduced by Gu and Soutchanski. Second, we have to restrict the Golog program to contain ground action only. Finally, we consider special classes of successor state axioms, namely the context-free ones and those that only admit local effects.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n Natural Language Processing in Domestic Service Robotics.\n \n \n \n\n\n \n Stefan Schiffer.\n\n\n \n\n\n\n In Stella Neumann; Thomas Niehr; Jens Runkehl; Paula Niemietz; and Jennifer Fest., editor(s),
LingUnite – Tag der Sprachforschung, Oct 11 2013. \n
Best Poster Award\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings { Schiffer_LingUnite2013_NLPinDSR,\n title = {Natural Language Processing in Domestic Service Robotics},\n booktitle = {LingUnite -- Tag der Sprachforschung},\n year = {2013},\n note = {Best Poster Award},\n month = {Oct 11},\n abstract = {As robots are more and more entering our everyday life, like as assistive devices to support us in our home with the daily chores, methods to control and to interact with such robots become more and more important. The most natural way for a human to instruct a robot is perhaps natural language. However, there are several challenges to master to allow for suitable humanrobot interaction by means of natural language. We will report on two of our efforts in enabling humans to use natural language to command a domestic service robot. The two methods we present reside on different levels - one is at a lower level of recognizing speech from acoustic input while the second one is about interpreting natural language. While the former was primarily intended for noisy scenarios to help rejecting utterances that were not meant for the robot, the latter yields a flexible system for commanding a robot which can resolve ambiguities and which is also capable of initiating steps to achieve clarification. The first approach [1] is at the signal processing stage where the acoustic input received from spoken language is to be converted to the textual level. When acting in human environments it is important that commands given to the robot are recognized robustly. Also, spoken language not directed to the robot must not be matched to an instruction for the robot to execute. We developed a system that is robust in noisy environments and that is insusceptible to act upon commands not meant for the robot. First, we use a threshold-based close speech detection to segment utterances targeted at the robot from the continuous audio stream recorded by a microphone. Then, we decode these utterances with two different decoders in parallel, namely one very restrictive decoder based on finite state grammars and a second more lenient decoder using N-grams. We do this to filter out false positive recognitions by comparing the output of the two decoders and rejecting the input if it was not recognized by both decoders. The second approach [2] takes place on a higher level of abstraction, that is, it deals with interpreting an utterance that has already been transformed to text from the raw audio signal. We model the processing of natural spoken language input as an interpretation process where the utterance needs to be mapped to a robot's capabilities. More precisely, we first analyse the given utterance syntactically by using a generic grammar that we developed for english directives. Then, we cast the interpretation as a planning problem where the individual actions available to the planner are to interpret syntactical elements of the utterance. If, in the course of interpreting, ambiguities are detected, the system uses decision-theory to weigh different alternatives. The system is also able to initiate clarification to resolve ambiguities and to handle errors as to arrive at a successful command interpretation eventually. We show how we evaluated several versions of the system with multiple utterances of different complexity as well as with incomplete and erroneous requests.},\n author = {Schiffer, Stefan},\n editor = {Stella Neumann and Thomas Niehr and Jens Runkehl and Paula Niemietz and Jennifer Fest}\n}\n\n
\n
\n\n\n
\n As robots are more and more entering our everyday life, like as assistive devices to support us in our home with the daily chores, methods to control and to interact with such robots become more and more important. The most natural way for a human to instruct a robot is perhaps natural language. However, there are several challenges to master to allow for suitable humanrobot interaction by means of natural language. We will report on two of our efforts in enabling humans to use natural language to command a domestic service robot. The two methods we present reside on different levels - one is at a lower level of recognizing speech from acoustic input while the second one is about interpreting natural language. While the former was primarily intended for noisy scenarios to help rejecting utterances that were not meant for the robot, the latter yields a flexible system for commanding a robot which can resolve ambiguities and which is also capable of initiating steps to achieve clarification. The first approach [1] is at the signal processing stage where the acoustic input received from spoken language is to be converted to the textual level. When acting in human environments it is important that commands given to the robot are recognized robustly. Also, spoken language not directed to the robot must not be matched to an instruction for the robot to execute. We developed a system that is robust in noisy environments and that is insusceptible to act upon commands not meant for the robot. First, we use a threshold-based close speech detection to segment utterances targeted at the robot from the continuous audio stream recorded by a microphone. Then, we decode these utterances with two different decoders in parallel, namely one very restrictive decoder based on finite state grammars and a second more lenient decoder using N-grams. We do this to filter out false positive recognitions by comparing the output of the two decoders and rejecting the input if it was not recognized by both decoders. The second approach [2] takes place on a higher level of abstraction, that is, it deals with interpreting an utterance that has already been transformed to text from the raw audio signal. We model the processing of natural spoken language input as an interpretation process where the utterance needs to be mapped to a robot's capabilities. More precisely, we first analyse the given utterance syntactically by using a generic grammar that we developed for english directives. Then, we cast the interpretation as a planning problem where the individual actions available to the planner are to interpret syntactical elements of the utterance. If, in the course of interpreting, ambiguities are detected, the system uses decision-theory to weigh different alternatives. The system is also able to initiate clarification to resolve ambiguities and to handle errors as to arrive at a successful command interpretation eventually. We show how we evaluated several versions of the system with multiple utterances of different complexity as well as with incomplete and erroneous requests.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n .\n \n \n \n\n\n \n Stefan Schiffer; Niklas Hoppe; and Gerhard Lakemeyer.\n\n\n \n\n\n\n Volume 358 of Communications in Computer and Information Science. Natural Language Interpretation for an Interactive Service Robot in Domestic Domains, pages 39–53. Joaquim Filipe; and Ana Fred., editor(s). Springer Berlin Heidelberg, 2013.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inbook {Schiffer:Hoppe:Lakemeyer:ICAART2012BOOK:NatLangInterpretation,\n title = {Natural Language Interpretation for an Interactive Service Robot in Domestic Domains},\n booktitle = {Agents and Artificial Intelligence. Fourth International Conference, ICAART 2012, Vilamoura, Portugal, February 6-8, 2012. Revised Selected Papers.},\n series = {Communications in Computer and Information Science},\n volume = {358},\n year = {2013},\n pages = {39--53},\n publisher = {Springer Berlin Heidelberg},\n organization = {Springer Berlin Heidelberg},\n abstract = {In this paper, we propose a flexible system for robust natural language interpretation of spoken commands on a mobile robot in domestic service robotics applications. Existing language processing for instructing a mobile robot is often restricted by using a simple grammar where precisely pre-defined utterances are directly mapped to system calls. These approaches do not regard fallibility of human users and they only allow for binary processing of an utterance; either a command is part of the grammar and hence understood correctly, or it is not part of the grammar and gets rejected. We model the language processing as an interpretation process where the utterance needs to be mapped to the robot's capabilities. We do so by casting the processing as a (decision-theoretic) planning problem on interpretation actions. This allows for a flexible system that can resolve ambiguities and which is also capable of initiating steps to achieve clarification. We show how we evaluated several versions of the system with multiple utterances of different complexity as well as with incomplete and erroneous requests.},\n author = {Schiffer, Stefan and Niklas Hoppe and Lakemeyer, Gerhard},\n editor = {Joaquim Filipe and Ana Fred}\n}\n\n
\n
\n\n\n
\n In this paper, we propose a flexible system for robust natural language interpretation of spoken commands on a mobile robot in domestic service robotics applications. Existing language processing for instructing a mobile robot is often restricted by using a simple grammar where precisely pre-defined utterances are directly mapped to system calls. These approaches do not regard fallibility of human users and they only allow for binary processing of an utterance; either a command is part of the grammar and hence understood correctly, or it is not part of the grammar and gets rejected. We model the language processing as an interpretation process where the utterance needs to be mapped to the robot's capabilities. We do so by casting the processing as a (decision-theoretic) planning problem on interpretation actions. This allows for a flexible system that can resolve ambiguities and which is also capable of initiating steps to achieve clarification. We show how we evaluated several versions of the system with multiple utterances of different complexity as well as with incomplete and erroneous requests.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Spatio-Temporal Reasoning about Traffic Scenarios.\n \n \n \n \n\n\n \n Christoph Schwering; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In
Eleventh International Symposium on Logical Formalizations of Commonsense Reasoning, Cyprus, 05/2013 2013. \n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n \n paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings { Schwering:Commonsense2013,\n title = {Spatio-Temporal Reasoning about Traffic Scenarios},\n booktitle = {Eleventh International Symposium on Logical Formalizations of Commonsense Reasoning},\n year = {2013},\n month = {05/2013},\n address = {Cyprus},\n abstract = {A model of spatial relations between automobiles changing over time needs to trade off expressivity, computational complexity, type of measured data, and closeness to human cognition. We present a common sense theory for reasoning about distances between cars based on two temporal distances between pairs of cars, \\emph{net time gap} and \\emph{time to collision}. We give an axiomatization in the situation calculus which allows reasoning about car-to-car relations and how they are affected by time and acceleration. We also discuss experimental results in a plan recognition scenario.},\n url = {http://kbsg.rwth-aachen.de/~schwering/commonsense-2013.pdf},\n url_Paper = {https://kbsg.rwth-aachen.de/sites/kbsg/files/SchweringLakemeyer2013.pdf},\n author = {Christoph Schwering and Lakemeyer, Gerhard}\n}\n\n
\n
\n\n\n
\n A model of spatial relations between automobiles changing over time needs to trade off expressivity, computational complexity, type of measured data, and closeness to human cognition. We present a common sense theory for reasoning about distances between cars based on two temporal distances between pairs of cars, \\emphnet time gap and \\emphtime to collision. We give an axiomatization in the situation calculus which allows reasoning about car-to-car relations and how they are affected by time and acceleration. We also discuss experimental results in a plan recognition scenario.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n Towards Decidable Verification of Non-Terminating Golog Programs.\n \n \n \n\n\n \n Martin Liebenberg.\n\n\n \n\n\n\n Diploma thesis, Department of Computer Science, RWTH Aachen University, April 2013.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@thesis{Liebenberg2013,\n author = {Martin Liebenberg},\n title = {Towards Decidable Verification of Non-Terminating\n {G}olog Programs},\n school = {Department of Computer Science, RWTH Aachen\n University},\n year = {2013},\n month = apr,\n advisor = {Cla{\\ss}en, Jens},\n abstract = {Claßen and Lakemeyer recently introduced algorithms\n for the verification of temporal properties of\n non-terminating Golog programs, based on the\n first-order modal Situation Calculus variant ES, and\n regression-based reasoning. However, while Golog’s\n high expressiveness is a desirable feature, it also\n means that their verification procedures cannot be\n guaranteed to terminate in general. In this thesis,\n we address this problem by showing that, for a\n relevant subset, the verification of non-terminating\n Golog programs is indeed decidable, which is\n achieved by means of three restrictions. First, we\n use the ES variant of a decidable two-variable\n fragment of the Situation Calculus that was\n introduced by Gu and Soutchanski. Second, we have to\n restrict the Golog program to contain ground action\n only. Finally, we consider special classes of\n successor state axioms, namely the context-free ones\n and those that only admit local effects.},\n type = {Diploma thesis}\n}\n\n
\n
\n\n\n
\n Claßen and Lakemeyer recently introduced algorithms for the verification of temporal properties of non-terminating Golog programs, based on the first-order modal Situation Calculus variant ES, and regression-based reasoning. However, while Golog’s high expressiveness is a desirable feature, it also means that their verification procedures cannot be guaranteed to terminate in general. In this thesis, we address this problem by showing that, for a relevant subset, the verification of non-terminating Golog programs is indeed decidable, which is achieved by means of three restrictions. First, we use the ES variant of a decidable two-variable fragment of the Situation Calculus that was introduced by Gu and Soutchanski. Second, we have to restrict the Golog program to contain ground action only. Finally, we consider special classes of successor state axioms, namely the context-free ones and those that only admit local effects.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n The Carologistics RoboCup Logistics Team 2013.\n \n \n \n\n\n \n Tim Niemueller; Daniel Ewert; Sebastian Reuter; Alexander Ferrein; Sabina Jeschke; and Gerhard Lakemeyer.\n\n\n \n\n\n\n Technical Report RWTH Aachen University and Aachen University of Applied Sciences, Aachen, Germany, June 2013.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@techreport{Carologistics2013,\n title =\t{{The Carologistics RoboCup Logistics Team 2013}},\n author =\t{Tim Niemueller and Daniel Ewert and Sebastian Reuter and\n Alexander Ferrein and Sabina Jeschke and Gerhard\n Lakemeyer},\n institution =\t{RWTH Aachen University and Aachen University of Applied Sciences},\n address =\t{Aachen, Germany},\n year =\t{2013},\n month =\t{June},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n Towards Benchmarking Cyber-Physical Systems in Factory Automation Scenarios.\n \n \n \n\n\n \n Tim Niemueller; Daniel Ewert; Sebastian Reuter; Ulrich Karras; Alexander Ferrein; Sabina Jeschke; and Gerhard Lakemeyer.\n\n\n \n\n\n\n In
KI 2013: Advances in Artificial Intelligence. Springer, 2013.\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@InCollection{LLSF2013,\n title={{Towards Benchmarking Cyber-Physical Systems in Factory\n Automation Scenarios}},\n author={Niemueller, Tim and Ewert, Daniel and Reuter, Sebastian\n and Karras, Ulrich and Ferrein, Alexander and Jeschke,\n Sabina and Lakemeyer, Gerhard},\n booktitle={KI 2013: Advances in Artificial Intelligence},\n year={2013},\n publisher={Springer}\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n Towards Deliberative Active Perception using Persistent Memory.\n \n \n \n\n\n \n Tim Niemueller; Nichola Abdo; Andreas Hertle; Gerhard Lakemeyer; Wolfram Burgard; and Bernhard Nebel.\n\n\n \n\n\n\n In
Proceedings of the Workshop on AI-based Robotics at the International Conference on Intelligent Robots and Systems (IROS), Tokio, Japan, 2013. \n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@InProceedings{Hybris-C1-Baseline,\n author={Tim Niemueller and Nichola Abdo and Andreas Hertle and Gerhard Lakemeyer and \n Wolfram Burgard and Bernhard Nebel},\n title={{Towards Deliberative Active Perception using Persistent\n Memory}},\n booktitle = {Proceedings of the Workshop on AI-based Robotics at the\n International Conference on Intelligent Robots and Systems (IROS)},\n year = 2013,\n\t\taddress = {Tokio, Japan}\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Incremental Task-level Reasoning in a Competitive Factory Automation Scenario.\n \n \n \n \n\n\n \n Tim Niemueller; Gerhard Lakemeyer; and Alexander Ferrein.\n\n\n \n\n\n\n In
AAAI Spring Symposium 2013 on Designing Intelligent Robots: Reintegrating AI II, Stanford, USA, March 25-27 2013. AAAI, AAAI\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@InProceedings {CLIPS-Agent,\n title = {Incremental Task-level Reasoning in a Competitive Factory Automation Scenario},\n booktitle = {AAAI Spring Symposium 2013 on Designing Intelligent Robots: Reintegrating AI II},\n year = {2013},\n month = {March 25-27},\n publisher = {AAAI},\n organization = {AAAI},\n address = {Stanford, USA},\n url = {https://www.fawkesrobotics.org/projects/clips-agent/},\n attachments = {https://kbsg.rwth-aachen.de/sites/kbsg/files/clips-agent-llsf-aaai-springsymp2013.pdf},\n author = {Tim Niemueller and Lakemeyer, Gerhard and Ferrein, Alexander}\n}\n\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n Proposal for Advancements to the LLSF in 2014 and beyond.\n \n \n \n\n\n \n Tim Niemueller; Gerhard Lakemeyer; Alexander Ferrein; Sebastian Reuter; Daniel Ewert; Sabina Jeschke; Dirk Pensky; and Ulrich Karras.\n\n\n \n\n\n\n In
ICAR – 1st Workshop on Developments in RoboCup Leagues, Montevideo, Uruguay, 2013. \n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@InProceedings{LLSF2014,\n title = {{Proposal for Advancements to the LLSF in 2014 and beyond}},\n author = {Tim Niemueller and Gerhard Lakemeyer and Alexander Ferrein\n and Sebastian Reuter and Daniel Ewert and Sabina\n Jeschke and Dirk Pensky and Ulrich Karras},\n booktitle = {ICAR -- 1st Workshop on Developments in RoboCup\n Leagues},\n year = 2013,\n address = {Montevideo, Uruguay},\n}\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n