Pub Date : 1996-10-09DOI: 10.1109/EURBOT.1996.552019
M. Ribeiro, J. Gonçalves
This paper presents a localisation procedure for a mobile robot operating in a structured environment whose map is totally or partially known a priori. The proposed approach is based on the recognition of natural landmarks identified through laser range data. With a rough estimate of the robot's position and orientation and the knowledge of an a priori map, the proposed technique chooses two vertical edges toward which, and using correctly chosen scanning parameters, a laser scan is directed. The post-processing of the so acquired range profile results in an updated estimate of the robot's location. The paper describes the different steps of the localisation technique, derives a criterion for the optimal choice of the pair of vertical edges, discusses relevant experimental results and presents directions for further research.
{"title":"Natural landmark based localisation of mobile robots using laser range data","authors":"M. Ribeiro, J. Gonçalves","doi":"10.1109/EURBOT.1996.552019","DOIUrl":"https://doi.org/10.1109/EURBOT.1996.552019","url":null,"abstract":"This paper presents a localisation procedure for a mobile robot operating in a structured environment whose map is totally or partially known a priori. The proposed approach is based on the recognition of natural landmarks identified through laser range data. With a rough estimate of the robot's position and orientation and the knowledge of an a priori map, the proposed technique chooses two vertical edges toward which, and using correctly chosen scanning parameters, a laser scan is directed. The post-processing of the so acquired range profile results in an updated estimate of the robot's location. The paper describes the different steps of the localisation technique, derives a criterion for the optimal choice of the pair of vertical edges, discusses relevant experimental results and presents directions for further research.","PeriodicalId":136786,"journal":{"name":"Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96)","volume":"516 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133068406","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 : 1996-10-09DOI: 10.1109/EURBOT.1996.551888
M. Piasecki
We have proposed a novel approach to perception of environment regularities which can be used by reactive navigation systems for mobile robots. The approach is inspired by existing eco-systems and the way how people and animals use them to recognise long term life conditions. The system provides the distributed model of the world which is build up in behavioural manner. We hope that the presented approach enables a more global composition and coordination of behaviours and at the same time it preserves the grace of reactive systems.
{"title":"Distributed perception for reactive navigation","authors":"M. Piasecki","doi":"10.1109/EURBOT.1996.551888","DOIUrl":"https://doi.org/10.1109/EURBOT.1996.551888","url":null,"abstract":"We have proposed a novel approach to perception of environment regularities which can be used by reactive navigation systems for mobile robots. The approach is inspired by existing eco-systems and the way how people and animals use them to recognise long term life conditions. The system provides the distributed model of the world which is build up in behavioural manner. We hope that the presented approach enables a more global composition and coordination of behaviours and at the same time it preserves the grace of reactive systems.","PeriodicalId":136786,"journal":{"name":"Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96)","volume":"133 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123717071","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 : 1996-10-09DOI: 10.1109/EURBOT.1996.551897
R. Cassinis, D. Grana, A. Rizzi
A localization system for mobile robots is presented. The system enables an autonomous mobile robot to locate itself along previously learned routes in a dynamic environment. The localization is based on data provided by an omnidirectional visual perception system employing optical pre-processing and neural network learning. New results have been achieved using colour information.
{"title":"Using colour information in an omnidirectional perception system for autonomous robot localization","authors":"R. Cassinis, D. Grana, A. Rizzi","doi":"10.1109/EURBOT.1996.551897","DOIUrl":"https://doi.org/10.1109/EURBOT.1996.551897","url":null,"abstract":"A localization system for mobile robots is presented. The system enables an autonomous mobile robot to locate itself along previously learned routes in a dynamic environment. The localization is based on data provided by an omnidirectional visual perception system employing optical pre-processing and neural network learning. New results have been achieved using colour information.","PeriodicalId":136786,"journal":{"name":"Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125263381","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 : 1996-10-09DOI: 10.1109/EURBOT.1996.551874
Wolfram Burgard, D. Fox, D. Hennig, Timo Schmidt
One of the main problems in the field of mobile robotics is the estimation of the robot's position in the environment. Position probability grids have been proven to be a robust technique for the estimation of the absolute position of a mobile robot. In this paper we describe an application of position probability grids to the tracking of the position of the robot. The main difference of our method to previous approaches lies in the fact that the position probability grid technique is a Bayesian approach which is able to deal with noisy sensors as well as ambiguities and is able to integrate sensor readings of different types of sensors over time. Given a starting position this method estimates the robot's current position by matching sensor readings against a metric model of the environment. Results described in this paper illustrate the robustness of this method against noisy sensors and errors in the environmental model.
{"title":"Position tracking with position probability grids","authors":"Wolfram Burgard, D. Fox, D. Hennig, Timo Schmidt","doi":"10.1109/EURBOT.1996.551874","DOIUrl":"https://doi.org/10.1109/EURBOT.1996.551874","url":null,"abstract":"One of the main problems in the field of mobile robotics is the estimation of the robot's position in the environment. Position probability grids have been proven to be a robust technique for the estimation of the absolute position of a mobile robot. In this paper we describe an application of position probability grids to the tracking of the position of the robot. The main difference of our method to previous approaches lies in the fact that the position probability grid technique is a Bayesian approach which is able to deal with noisy sensors as well as ambiguities and is able to integrate sensor readings of different types of sensors over time. Given a starting position this method estimates the robot's current position by matching sensor readings against a metric model of the environment. Results described in this paper illustrate the robustness of this method against noisy sensors and errors in the environmental model.","PeriodicalId":136786,"journal":{"name":"Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128839457","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 : 1996-10-01DOI: 10.1109/EURBOT.1996.551885
M. Bekambo, G. Deconinck, R. Lauwereins, J. Peperstraete, J. Vandorpe, H. Van Brussel
The triple modular redundant (TMR) architecture is based on the triplication of application modules To mask faults, copies of modules are mapped on processing units, capable of direct communication. In the approach used in this paper, each processing unit (or a processor) mapping a module, is equipped with a kernel. This kernel allows it to mask faults via an agreement protocol involving the two other processors (mapping the two other copies). Faults are masked, under user requests, through kernel primitives. Robot movements are observed when its path following controller is mapped on a TMR architecture. The output of the velocity controller is then submitted to an agreement. The study of resulted robot movements shows that single faults are efficiently masked. There is no perceptible difference among the robot behaviors, when using or not using fault-masking kernel, unless when processors faults occur.
{"title":"Validation of kernel-based TMR in an autonomous guided vehicle","authors":"M. Bekambo, G. Deconinck, R. Lauwereins, J. Peperstraete, J. Vandorpe, H. Van Brussel","doi":"10.1109/EURBOT.1996.551885","DOIUrl":"https://doi.org/10.1109/EURBOT.1996.551885","url":null,"abstract":"The triple modular redundant (TMR) architecture is based on the triplication of application modules To mask faults, copies of modules are mapped on processing units, capable of direct communication. In the approach used in this paper, each processing unit (or a processor) mapping a module, is equipped with a kernel. This kernel allows it to mask faults via an agreement protocol involving the two other processors (mapping the two other copies). Faults are masked, under user requests, through kernel primitives. Robot movements are observed when its path following controller is mapped on a TMR architecture. The output of the velocity controller is then submitted to an agreement. The study of resulted robot movements shows that single faults are efficiently masked. There is no perceptible difference among the robot behaviors, when using or not using fault-masking kernel, unless when processors faults occur.","PeriodicalId":136786,"journal":{"name":"Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128976185","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}