Pub Date : 2004-06-14DOI: 10.1109/IVS.2004.1336393
T. Dang, C. Hoffmann
In this paper we present a self-calibration approach that updates the extrinsic parameters and the focal lengths of a stereo vision sensor. We employ a recursive estimation algorithm based on an Extended Kalman Filter. To improve the self-calibration process, we introduce a robust innovation stage for the Kalman filter: A Least Median Squares estimator is employed to eliminate outliers and thus to achieve better performance. The algorithm gives promising results on experiments with synthetic and natural imagery.
{"title":"Stereo calibration in vehicles","authors":"T. Dang, C. Hoffmann","doi":"10.1109/IVS.2004.1336393","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336393","url":null,"abstract":"In this paper we present a self-calibration approach that updates the extrinsic parameters and the focal lengths of a stereo vision sensor. We employ a recursive estimation algorithm based on an Extended Kalman Filter. To improve the self-calibration process, we introduce a robust innovation stage for the Kalman filter: A Least Median Squares estimator is employed to eliminate outliers and thus to achieve better performance. The algorithm gives promising results on experiments with synthetic and natural imagery.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126487779","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336348
D. Gavrila, J. Giebel, S. Munder
This paper presents the results of the first large-scale field tests on vision-based pedestrian protection from a moving vehicle. Our PROTECTOR system combines pedestrian detection, trajectory estimation, risk assessment and driver warning. The paper pursues a "system approach" related to the detection component. An optimization scheme models the system as a succession of individual modules and finds a good overall parameter setting by combining individual ROCs using a convex-hull technique. On the experimental side, we present a methodology for the validation of the pedestrian detection performance in an actual vehicle setting. We hope this test methodology to contribute towards the establishment of benchmark testing, enabling this application to mature. We validate the PROTECTOR system using the proposed methodology and present interesting quantitative results based on tens of thousands of images from hours of driving. Although results are promising, more research is needed before such systems can be placed at the hands of ordinary vehicle drivers.
{"title":"Vision-based pedestrian detection: the PROTECTOR system","authors":"D. Gavrila, J. Giebel, S. Munder","doi":"10.1109/IVS.2004.1336348","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336348","url":null,"abstract":"This paper presents the results of the first large-scale field tests on vision-based pedestrian protection from a moving vehicle. Our PROTECTOR system combines pedestrian detection, trajectory estimation, risk assessment and driver warning. The paper pursues a \"system approach\" related to the detection component. An optimization scheme models the system as a succession of individual modules and finds a good overall parameter setting by combining individual ROCs using a convex-hull technique. On the experimental side, we present a methodology for the validation of the pedestrian detection performance in an actual vehicle setting. We hope this test methodology to contribute towards the establishment of benchmark testing, enabling this application to mature. We validate the PROTECTOR system using the proposed methodology and present interesting quantitative results based on tens of thousands of images from hours of driving. Although results are promising, more research is needed before such systems can be placed at the hands of ordinary vehicle drivers.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130167194","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336368
C. Nitsche, S. Schroedl, W. Weiss
Fuel cell vehicles and fuel cell research is one of the newer areas in automotive technology. This paper describes an approach that utilizes artificial neural networks to alleviate the task of onboard diagnostics for fuel cell vehicles. The basic idea is an online learning scenario that trains a power train model with every-day driving data; this model can then be used to estimate a characteristic curve by feeding it with predefined input variables corresponding to the constant conditions of a stationary workshop test. In this way, a major obstacle for on-line diagnosis, namely the multitude of varying nuisance variables, can be compensated for. For a diagnosis algorithm, it is considerably easier to compare the resulting predicted characteristic curve with an ideal reference curve, rather than to directly deal with all the influence factors.
{"title":"Onboard diagnostics concept for fuel cell vehicles using adaptive modelling","authors":"C. Nitsche, S. Schroedl, W. Weiss","doi":"10.1109/IVS.2004.1336368","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336368","url":null,"abstract":"Fuel cell vehicles and fuel cell research is one of the newer areas in automotive technology. This paper describes an approach that utilizes artificial neural networks to alleviate the task of onboard diagnostics for fuel cell vehicles. The basic idea is an online learning scenario that trains a power train model with every-day driving data; this model can then be used to estimate a characteristic curve by feeding it with predefined input variables corresponding to the constant conditions of a stationary workshop test. In this way, a major obstacle for on-line diagnosis, namely the multitude of varying nuisance variables, can be compensated for. For a diagnosis algorithm, it is considerably easier to compare the resulting predicted characteristic curve with an ideal reference curve, rather than to directly deal with all the influence factors.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127838287","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336458
A. Polychronopoulos, M. Tsogas, A. Amditis, U. Scheunert, L. Andreone, F. Tango
Situation and threat assessment is considered as the highest level of abstraction in the vehicle tracking processes. In this paper, a broad discussion is introduced on algorithms for active safety functions, whilst a new dynamic algorithm is proposed. This approach handles all objects' states as dynamic stochastic variables and based on a Kalman approach calculates in real time all trajectories respectively. Thus, a reconstruction of the traffic scene can be achieved in order to assess a level of threat for all moving and stationary obstacles in the longitudinal area of the subject vehicle. This approach is adopted in the European co-funded project "EUCLIDE", which develops a vision enhancement and collision warning system merging the functionality of an infrared camera and mmw radar sensor. Results are presented using simulated and real data sets from dedicated sessions.
{"title":"Dynamic situation and threat assessment for collision warning systems: the EUCLIDE approach","authors":"A. Polychronopoulos, M. Tsogas, A. Amditis, U. Scheunert, L. Andreone, F. Tango","doi":"10.1109/IVS.2004.1336458","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336458","url":null,"abstract":"Situation and threat assessment is considered as the highest level of abstraction in the vehicle tracking processes. In this paper, a broad discussion is introduced on algorithms for active safety functions, whilst a new dynamic algorithm is proposed. This approach handles all objects' states as dynamic stochastic variables and based on a Kalman approach calculates in real time all trajectories respectively. Thus, a reconstruction of the traffic scene can be achieved in order to assess a level of threat for all moving and stationary obstacles in the longitudinal area of the subject vehicle. This approach is adopted in the European co-funded project \"EUCLIDE\", which develops a vision enhancement and collision warning system merging the functionality of an infrared camera and mmw radar sensor. Results are presented using simulated and real data sets from dedicated sessions.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121238578","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336425
Wuhong Wang, Wei Zhang, D. Li, K. Hirahara, K. Ikeuchi
Car-following modelling in traffic flow theory has been becoming of increasing importance in traffic engineering and Intelligent Transport System(ITS), the point of concentration in this research field is how to analysis and measurement of driver cognitive behaviour. Based on qualitative description of driving behaviour with the new concept of driver's multi-typed information process and multi-ruled decision-making mechanism, this paper has analysed in more detail the AP (action point) model, and ameliorated AP model by eliminating its deficiency. The emphasis of this paper is placed on the deduction of the acceleration equations by considering that the following car is subjected in congested traffic flow. Furthermore, from the cybernetics perspective, this paper has carried out numeral simulation to car-following behaviour with deceleration and acceleration algorithms. The model validation and simulation results have shown that the improved action point car-following model can replicated car-following behaviour and be able to use to reveal the essence of traffic flow characteristics.
{"title":"Improved action point model in traffic flow based on driver's cognitive mechanism","authors":"Wuhong Wang, Wei Zhang, D. Li, K. Hirahara, K. Ikeuchi","doi":"10.1109/IVS.2004.1336425","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336425","url":null,"abstract":"Car-following modelling in traffic flow theory has been becoming of increasing importance in traffic engineering and Intelligent Transport System(ITS), the point of concentration in this research field is how to analysis and measurement of driver cognitive behaviour. Based on qualitative description of driving behaviour with the new concept of driver's multi-typed information process and multi-ruled decision-making mechanism, this paper has analysed in more detail the AP (action point) model, and ameliorated AP model by eliminating its deficiency. The emphasis of this paper is placed on the deduction of the acceleration equations by considering that the following car is subjected in congested traffic flow. Furthermore, from the cybernetics perspective, this paper has carried out numeral simulation to car-following behaviour with deceleration and acceleration algorithms. The model validation and simulation results have shown that the improved action point car-following model can replicated car-following behaviour and be able to use to reveal the essence of traffic flow characteristics.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129306141","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336352
E. Dagan, O. Mano, G. Stein, A. Shashua
The large number of rear end collisions due to driver inattention has been identified as a major automotive safety issue. Even a short advance warning can significantly reduce the number and severity of the collisions. This paper describes a vision based forward collision warning (FCW) system for highway safety. The algorithm described in this paper computes time to contact (TTC) and possible collision course directly from the size and position of the vehicles in the image - which are the natural measurements for a vision based system - without having to compute a 3D representation of the scene. The use of a single low cost image sensor results in an affordable system which is simple to install. The system has been implemented on real-time hardware and has been test driven on highways. Collision avoidance tests have also been performed on test tracks.
{"title":"Forward collision warning with a single camera","authors":"E. Dagan, O. Mano, G. Stein, A. Shashua","doi":"10.1109/IVS.2004.1336352","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336352","url":null,"abstract":"The large number of rear end collisions due to driver inattention has been identified as a major automotive safety issue. Even a short advance warning can significantly reduce the number and severity of the collisions. This paper describes a vision based forward collision warning (FCW) system for highway safety. The algorithm described in this paper computes time to contact (TTC) and possible collision course directly from the size and position of the vehicles in the image - which are the natural measurements for a vision based system - without having to compute a 3D representation of the scene. The use of a single low cost image sensor results in an affordable system which is simple to install. The system has been implemented on real-time hardware and has been test driven on highways. Collision avoidance tests have also been performed on test tracks.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129278273","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}
Implementation of the Advanced Cruise-assist Highway System requires rigorous testing of the road sensors, which play a central role in the system, on actual roads to ascertain their vehicle detection characteristics. We evaluated these detection characteristics on actual roads under conditions of poor visibility caused by fog. This report presents an overview of the test results and issues raised for operational deployment of the system.
{"title":"Evaluation of the detection characteristics of road sensors under poor-visibility conditions","authors":"Ryoichi Kurata, Hideki Watanabe, Masatoshi Tohno, Takakazu Ishii, Hiroyuki Oouchi","doi":"10.1109/IVS.2004.1336441","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336441","url":null,"abstract":"Implementation of the Advanced Cruise-assist Highway System requires rigorous testing of the road sensors, which play a central role in the system, on actual roads to ascertain their vehicle detection characteristics. We evaluated these detection characteristics on actual roads under conditions of poor visibility caused by fog. This report presents an overview of the test results and issues raised for operational deployment of the system.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116730562","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336471
T. Korah, C. Rasmussen
Over the past few years, global positioning systems (GPS) have been increasingly used in passenger and commercial vehicles for navigation and vehicle tracking purposes. In practice, GPS systems are prone to systematic errors and intermittent drop-outs that degrade the accuracy of the sensor. In this work, we describe an approach to localizing vehicles with respect to the road given erroneous sensor measurements using only aerial images. Our method works on both urban and rural areas, while being robust to a number of occlusions and shadows. The spatial tracker incorporates multiple measurement models with varying constraints, automatically detecting and switching to the appropriate model. We demonstrate our technique by correcting in real-time highly inaccurate GPS readings collected while driving in diverse areas.
{"title":"Probabilistic contour extraction with model-switching for vehicle localization","authors":"T. Korah, C. Rasmussen","doi":"10.1109/IVS.2004.1336471","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336471","url":null,"abstract":"Over the past few years, global positioning systems (GPS) have been increasingly used in passenger and commercial vehicles for navigation and vehicle tracking purposes. In practice, GPS systems are prone to systematic errors and intermittent drop-outs that degrade the accuracy of the sensor. In this work, we describe an approach to localizing vehicles with respect to the road given erroneous sensor measurements using only aerial images. Our method works on both urban and rural areas, while being robust to a number of occlusions and shadows. The spatial tracker incorporates multiple measurement models with varying constraints, automatically detecting and switching to the appropriate model. We demonstrate our technique by correcting in real-time highly inaccurate GPS readings collected while driving in diverse areas.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114256945","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336384
M. Saniee, J. Habibi
Real-time, fuzzy rule-based guidance systems for autonomous vehicles on limited-access highways are investigated. The goal of these systems is to plan trajectories that are safe, while satisfying the driver's requests based on stochastic information about the vehicle's state and the surrounding traffic. This paper presented a new method to implement an automated highway driving behaviour. The main advantage of the suggested system is its well-defined structure. To test the designed system, a simulation tool is implemented. By using the described tool, we can analyse the operation of the implemented decision making system in a simulated highway. Results show an acceptable performance of the developed fuzzy system.
{"title":"A fuzzy ranking method for automated highway driving","authors":"M. Saniee, J. Habibi","doi":"10.1109/IVS.2004.1336384","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336384","url":null,"abstract":"Real-time, fuzzy rule-based guidance systems for autonomous vehicles on limited-access highways are investigated. The goal of these systems is to plan trajectories that are safe, while satisfying the driver's requests based on stochastic information about the vehicle's state and the surrounding traffic. This paper presented a new method to implement an automated highway driving behaviour. The main advantage of the suggested system is its well-defined structure. To test the designed system, a simulation tool is implemented. By using the described tool, we can analyse the operation of the implemented decision making system in a simulated highway. Results show an acceptable performance of the developed fuzzy system.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114277704","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 : 2004-06-14DOI: 10.1109/IVS.2004.1336491
N. Kaempchen, K. Weiß, M. Schaefer, K. Dietmayer
Classical object tracking approaches use a Kalman-filter with a single dynamic model which is therefore optimised to a single driving maneuver. In contrast the interacting multiple model (IMM) filter allows for several parallel models which are combined to a weighted estimate. Choosing models for different driving modes, such as constant speed, acceleration and strong acceleration changes, the object state estimation can be optimised for highly dynamic driving maneuvers. The paper describes the analysis of Stop&Go situations and the systematic parametrisation of the IMM method based on these statistics. The evaluation of the IMM approach is presented based on real sensor measurements of laser scanners, a radar and a video image processing unit.
{"title":"IMM object tracking for high dynamic driving maneuvers","authors":"N. Kaempchen, K. Weiß, M. Schaefer, K. Dietmayer","doi":"10.1109/IVS.2004.1336491","DOIUrl":"https://doi.org/10.1109/IVS.2004.1336491","url":null,"abstract":"Classical object tracking approaches use a Kalman-filter with a single dynamic model which is therefore optimised to a single driving maneuver. In contrast the interacting multiple model (IMM) filter allows for several parallel models which are combined to a weighted estimate. Choosing models for different driving modes, such as constant speed, acceleration and strong acceleration changes, the object state estimation can be optimised for highly dynamic driving maneuvers. The paper describes the analysis of Stop&Go situations and the systematic parametrisation of the IMM method based on these statistics. The evaluation of the IMM approach is presented based on real sensor measurements of laser scanners, a radar and a video image processing unit.","PeriodicalId":296386,"journal":{"name":"IEEE Intelligent Vehicles Symposium, 2004","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127422953","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：