Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827641
M. Kasper, Gernot Fricke, E. Puttkamer
The article gives an overview of learning aspects regarding behavior based control architectures for autonomous mobile robots. We propose a common, modularized architecture, which is particularly suited for learning experiments. The main focus lies in "Learning from Demonstration" in a spatial domain, which means teaching motor behaviors by humans or other robots. First results applying RBF approximation, growing neural cell structures and probabilistic models for progress estimation are presented.
{"title":"A behavior-based architecture for teaching more than reactive behaviors to mobile robots","authors":"M. Kasper, Gernot Fricke, E. Puttkamer","doi":"10.1109/EURBOT.1999.827641","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827641","url":null,"abstract":"The article gives an overview of learning aspects regarding behavior based control architectures for autonomous mobile robots. We propose a common, modularized architecture, which is particularly suited for learning experiments. The main focus lies in \"Learning from Demonstration\" in a spatial domain, which means teaching motor behaviors by humans or other robots. First results applying RBF approximation, growing neural cell structures and probabilistic models for progress estimation are presented.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131374440","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827627
A. Chella, M. Frixione, S. Gaglio
An autonomous robot involved in long and complex missions should be able to generate, update and process its own plans of action. In this perspective, it is not plausible for the meaning of the representations used by the robot to be given from outside the system itself. Rather, the meaning of internal symbols must be firmly anchored to the world through the perceptual abilities and the overall activities of the robot. According to these premises, we present an approach to action representation that is based on a "conceptual" level of representation, acting as an intermediate level between the symbols and data coming from the sensors. Symbolic representations are interpreted by mapping them onto the conceptual level through a mapping mechanism based on artificial neural networks. Examples of the proposed framework are reported, based on experiments performed on a RWI-B12 autonomous robot.
{"title":"A conceptual representation of the actions of an autonomous robot","authors":"A. Chella, M. Frixione, S. Gaglio","doi":"10.1109/EURBOT.1999.827627","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827627","url":null,"abstract":"An autonomous robot involved in long and complex missions should be able to generate, update and process its own plans of action. In this perspective, it is not plausible for the meaning of the representations used by the robot to be given from outside the system itself. Rather, the meaning of internal symbols must be firmly anchored to the world through the perceptual abilities and the overall activities of the robot. According to these premises, we present an approach to action representation that is based on a \"conceptual\" level of representation, acting as an intermediate level between the symbols and data coming from the sensors. Symbolic representations are interpreted by mapping them onto the conceptual level through a mapping mechanism based on artificial neural networks. Examples of the proposed framework are reported, based on experiments performed on a RWI-B12 autonomous robot.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"207 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131920086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827630
S. Enderle, G. Kraetzschmar, S. Sablatnog, G. Palm
Sensor interpretation in mobile robots often involves an inverse sensor model, which generates hypotheses on specific aspects of the robot's environment based on current sensor data. Building inverse sensor models for sonar sensor assemblies is a particularly difficult problem that has received much attention in the past few years. A common solution is to train neural networks using supervised learning. However, large amounts of training data are typically needed, consisting, for example, of scans of recorded sonar data which are labeled with manually constructed teacher maps. Obtaining these training data is an error-prone and time-consuming process. We suggest that it can be avoided if an additional sensor, like a laser scanner, is also available which can act as the feeding signal. We have successfully trained inverse sensor models for sonar interpretation using laser scan data. In this paper, we describe the procedure we used and the results we obtained.
{"title":"Sonar interpretation learned from laser data","authors":"S. Enderle, G. Kraetzschmar, S. Sablatnog, G. Palm","doi":"10.1109/EURBOT.1999.827630","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827630","url":null,"abstract":"Sensor interpretation in mobile robots often involves an inverse sensor model, which generates hypotheses on specific aspects of the robot's environment based on current sensor data. Building inverse sensor models for sonar sensor assemblies is a particularly difficult problem that has received much attention in the past few years. A common solution is to train neural networks using supervised learning. However, large amounts of training data are typically needed, consisting, for example, of scans of recorded sonar data which are labeled with manually constructed teacher maps. Obtaining these training data is an error-prone and time-consuming process. We suggest that it can be avoided if an additional sensor, like a laser scanner, is also available which can act as the feeding signal. We have successfully trained inverse sensor models for sonar interpretation using laser scan data. In this paper, we describe the procedure we used and the results we obtained.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"947 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116435920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827640
C. Schlegel, R. Worz
Advanced mobile robots have to cope with many different situations and have to fulfil their tasks even in dynamic environments. Furthermore, there is an increasing demand that those systems perform not only a single task but show a whole set of different capabilities. This includes the concurrent and interruptable execution of several tasks. Reliably performing different tasks over long periods of time not only requires advanced basic skills, but also an appropriate robot control architecture. This includes, e.g., mechanisms to coordinate the execution of competing tasks. We present the architecture used on our B21 demonstrator of the SFB 527. A special emphasis is laid on aspects of our supporting software framework, SMARTSOFT, and the implemented set of skills.
{"title":"Interfacing different layers of a multilayer architecture for sensorimotor systems using the object-oriented framework SMARTSOFT","authors":"C. Schlegel, R. Worz","doi":"10.1109/EURBOT.1999.827640","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827640","url":null,"abstract":"Advanced mobile robots have to cope with many different situations and have to fulfil their tasks even in dynamic environments. Furthermore, there is an increasing demand that those systems perform not only a single task but show a whole set of different capabilities. This includes the concurrent and interruptable execution of several tasks. Reliably performing different tasks over long periods of time not only requires advanced basic skills, but also an appropriate robot control architecture. This includes, e.g., mechanisms to coordinate the execution of competing tasks. We present the architecture used on our B21 demonstrator of the SFB 527. A special emphasis is laid on aspects of our supporting software framework, SMARTSOFT, and the implemented set of skills.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"206 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121849030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827624
E. Meier, F. Ade
The detection of objects in every frame of a sequence is often not sufficient for scene interpretation. Tracking can increase the robustness, especially when occlusions occur or when objects temporally disappear. The standard approach for tracking is to use a Kalman filter for every object. This, however requires the use of a high complexity management system to deal with the multiple hypotheses necessary to track all anticipated objects. We present a stochastic approach which is based on the CONDENSATION algorithm-conditional density propagation over time-that is capable of tracking multiple objects with multiple hypotheses in range images. A probability density function describing the likely state of the objects is propagated over time using a dynamic model. The measurements influence the probability function and allow the incorporation of new objects into the tracking scheme. Additionally, the representation of the density function with a fixed number of samples ensures a constant running time per iteration step. Results on different data sources are shown for mobile robot applications.
{"title":"Using the condensation algorithm to implement tracking for mobile robots","authors":"E. Meier, F. Ade","doi":"10.1109/EURBOT.1999.827624","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827624","url":null,"abstract":"The detection of objects in every frame of a sequence is often not sufficient for scene interpretation. Tracking can increase the robustness, especially when occlusions occur or when objects temporally disappear. The standard approach for tracking is to use a Kalman filter for every object. This, however requires the use of a high complexity management system to deal with the multiple hypotheses necessary to track all anticipated objects. We present a stochastic approach which is based on the CONDENSATION algorithm-conditional density propagation over time-that is capable of tracking multiple objects with multiple hypotheses in range images. A probability density function describing the likely state of the objects is propagated over time using a dynamic model. The measurements influence the probability function and allow the incorporation of new objects into the tracking scheme. Additionally, the representation of the density function with a fixed number of samples ensures a constant running time per iteration step. Results on different data sources are shown for mobile robot applications.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117307477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827644
N. Tschichold-Gurman, S. J. Vestli, G. Schweitzer
The Mobile Post Distribution system MOPS has been developed at ETH Zurich over the last few years (to 1999). Numerous technical problems had to be solved during this period. Currently, we are operating and testing it in our office building. A service robot like MOPS is a highly complex system, as it includes, in addition to the robot itself, the task, the operator, and the environment. Operating experiences with such service robots are not easily available yet. They would be of value, as in the end customer acceptance will decide upon the actual use of service robots. Technical aspects of the realization of MOPS are discussed briefly. Furthermore, operating experiences, lessons learned and consequences are commented upon.
{"title":"Operating experiences with the service robot MOPS","authors":"N. Tschichold-Gurman, S. J. Vestli, G. Schweitzer","doi":"10.1109/EURBOT.1999.827644","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827644","url":null,"abstract":"The Mobile Post Distribution system MOPS has been developed at ETH Zurich over the last few years (to 1999). Numerous technical problems had to be solved during this period. Currently, we are operating and testing it in our office building. A service robot like MOPS is a highly complex system, as it includes, in addition to the robot itself, the task, the operator, and the environment. Operating experiences with such service robots are not easily available yet. They would be of value, as in the end customer acceptance will decide upon the actual use of service robots. Technical aspects of the realization of MOPS are discussed briefly. Furthermore, operating experiences, lessons learned and consequences are commented upon.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117002293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827634
M. Piaggio, A. Sgorbissa, R. Zaccaria
A novel approach to the global localisation problem for an autonomous mobile robot is presented. Instead of referring to traditional map-based techniques, we choose to extract a graph-like topological representation of the free space from occupancy grids, thus shifting the map-matching problem to a sub-graph isomorphism one. An efficient any-time algorithm is described in detail, and simulated experimental results are provided.
{"title":"Global localisation via sub-graph isomorphism","authors":"M. Piaggio, A. Sgorbissa, R. Zaccaria","doi":"10.1109/EURBOT.1999.827634","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827634","url":null,"abstract":"A novel approach to the global localisation problem for an autonomous mobile robot is presented. Instead of referring to traditional map-based techniques, we choose to extract a graph-like topological representation of the free space from occupancy grids, thus shifting the map-matching problem to a sub-graph isomorphism one. An efficient any-time algorithm is described in detail, and simulated experimental results are provided.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128082449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827620
P. Skrzypczyński, P. Drapikowski
The perception and world model building methods for multiple mobile robots operating in an industrial environment are discussed. The concept of geometrical database is associated with multiple classes of perception-based maps, to obtain the world model. Task-oriented maps are produced by grouped sensors. Two different representations of the environment are considered, namely vector-based and raster-based maps. Sensed maps are checked for consistency with the predefined model. The multi-agent approach has been applied with respect to the system architecture. As a consequence, world-model maintenance activity consists of a number of processes, which are driven by the team of agents. The results of experimental map building by using data from the optical range sensor combined with the predefined model are presented.
{"title":"Environment modelling in a multi-agent mobile system","authors":"P. Skrzypczyński, P. Drapikowski","doi":"10.1109/EURBOT.1999.827620","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827620","url":null,"abstract":"The perception and world model building methods for multiple mobile robots operating in an industrial environment are discussed. The concept of geometrical database is associated with multiple classes of perception-based maps, to obtain the world model. Task-oriented maps are produced by grouped sensors. Two different representations of the environment are considered, namely vector-based and raster-based maps. Sensed maps are checked for consistency with the predefined model. The multi-agent approach has been applied with respect to the system architecture. As a consequence, world-model maintenance activity consists of a number of processes, which are driven by the team of agents. The results of experimental map building by using data from the optical range sensor combined with the predefined model are presented.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114660896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827635
J. Crowley
The study of mobile robotics emerged as a scientific discipline with experiments on path planning and navigation on the Shakey system at SRI. The field exploded with interest in the early 1 9 8 0 ’ ~ ~ inspired in part by funding from the US DARPA Autonomous Vehicle program. By the late 1980’s the field had sufficiently blossomed to take an important role in robotics science, as evident from the number of papers in conferences and journals. Many techniques reported in scientific conferences in the late 90’s are modem instantiations of techniques first reported in the mid and late 1980’s.
{"title":"20 years of progress in mobile robotics: a historical perspective","authors":"J. Crowley","doi":"10.1109/EURBOT.1999.827635","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827635","url":null,"abstract":"The study of mobile robotics emerged as a scientific discipline with experiments on path planning and navigation on the Shakey system at SRI. The field exploded with interest in the early 1 9 8 0 ’ ~ ~ inspired in part by funding from the US DARPA Autonomous Vehicle program. By the late 1980’s the field had sufficiently blossomed to take an important role in robotics science, as evident from the number of papers in conferences and journals. Many techniques reported in scientific conferences in the late 90’s are modem instantiations of techniques first reported in the mid and late 1980’s.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125402424","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1999-09-06DOI: 10.1109/EURBOT.1999.827633
G. Bianco, A. Rizzi, R. Cassinis, N. Adami
Analyses the guidance principle and the robustness features of a biologically-inspired visual homing algorithm for autonomous robots. The homing strategy uses colour images and is based on an affine matching model whose parameters are used to estimate real navigation displacement in the environment. The guidance principle of the visual homing is proven to be a visual potential function with an equilibrium point located at the goal position. The presence of a potential function means that classical control-theory principles based on the Lyapunov functions can be applied to assess the robustness of the navigation strategy.
{"title":"Guidance principle and robustness issues for a biologically-inspired visual homing","authors":"G. Bianco, A. Rizzi, R. Cassinis, N. Adami","doi":"10.1109/EURBOT.1999.827633","DOIUrl":"https://doi.org/10.1109/EURBOT.1999.827633","url":null,"abstract":"Analyses the guidance principle and the robustness features of a biologically-inspired visual homing algorithm for autonomous robots. The homing strategy uses colour images and is based on an affine matching model whose parameters are used to estimate real navigation displacement in the environment. The guidance principle of the visual homing is proven to be a visual potential function with an equilibrium point located at the goal position. The presence of a potential function means that classical control-theory principles based on the Lyapunov functions can be applied to assess the robustness of the navigation strategy.","PeriodicalId":364500,"journal":{"name":"1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355)","volume":"PP 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126447789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：