L. Iocchi and D. Nardi. Self-localization in the RoboCup environment. In RoboCup-99: Robot Soccer World Cup III, pages 318-330, 1999. 24  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the camera with the 3D model of the world. Two approaches are possible: matching in image space and matching in world space; 18] surveys both methods. If the camera pose and internal parameters are known, features from one space can be projected into the other space. When matching in world space [3,12,17] the movement of the world projected image features is directly determined by the camera pose, but the error in the projected image features is dependent on the distance to the camera. Furthermore, the image features have to be found first, which is usually costly. When matching in image space, ....

L. Iocchi, D. Nardi, Self-localization in the RoboCup environment, Proceedings of the Third International Workshop on RoboCup (1999) 115.

....(self localization) To recover the pose of the robot we match a model of the world with the observed camera imagW The match takes place in imag space or in world space. In [1] a survey of both methods of Model to Imag Reg stration is g ven. In RoboCup, matching is mostly done in world space [2,3,4,5] because the movement of the features in world space is directly determined by the movement of the robot. However, the error in the feature positions in world space is dependent on the distance to the camera. When matching is done in imag e space, the robot pose is more di#cult to describeusing ....

Luca Iocchi and Daniele Nardi, "Self-Localization in the RoboCup Environment", Proc. Third International Workshop on RoboCup, p115

....differ from this approach mainly by using vision data instead of laser data. The main challenge for vision algorithms is, that they have to cope with the larger amount of data. Further approaches to vision based self localization using directional and omnidirectional cameras can be found in [20] [21], 22] 23] 24] Most of them are data driven, e.g. apply a Hough transformation or other computational expensive feature extraction technique to the complete image, whereas our approach is model driven and requires only the evaluation of a few pixels in the vicinity of a projected model ....

L. Iocchi and D. Nardi, "Self-Localization in the RoboCup Environment," in Third International Workshop on RoboCup (Robot World Cup Soccer Games and Conferences). 1999, Lecture Notes in Computer Science, Springer-Verlag.

....alternatively, range scanners, based on sonar or laser, can be used for the detection of walls [7] Environments based on a line segment description are easy to build manually and can describe quite well several environments. Line extraction is often accomplished through the Hough Transform [8] [9] [10] but many other approaches can be found in the literature [11] 12] In order to overcome the difculties related to natural feature recognition, some researchers proposed to use beacons, i.e. special patterns that can be more easily recognized. Although their use simplies the recognition ....

Iocchi, L., Nardi, D.: Self-localization in the robocup environment. In: proceedings of the International Workshop on RoboCup. (1999) 318--330

....which are very accurate in depth estimation, they get away with a simpler state estimation mechanism in which can assume almost perfect sensors with known positions. Most other mid size teams coordinate the play of their team mates by negotiating or assigning roles to the different players [8]. In contrast, in the AGILO team the coordination is implicit and based on a sophisticated cost estimate for task assignment. The AGILO team is also distinguished in the mid size league with respect to its extensive use of learning and plan based control mechanisms. Technologically, the AGILO ....

L. Iocchi and D. Nardi. Self-Localization in the RoboCup Environment. In Third International Workshop on RoboCup (Robot World Cup Soccer Games and Conferences), Lecture Notes in Computer Science. Springer-Verlag, 1999.

....competition in 2000 in Amsterdam. 1 Introduction The ability to determine its global position is crucial for a robot to succeed in the RoboCup environment. This is especially true for the goal keeper robot. So there have been some publications on localization for RoboCup robots already ( 5] [6], 2] One of these approaches relies on the very precise measurement of distances with a laser scanner. The scan is segmented in lines, which are matched against an a priori model of the environment. The drawback of this method is the use of a laser scanner, which is an expensive sensor in terms ....

Iocchi L., Nardi D., Self-Localization in the RoboCup Environment, RoboCup-99: The Third Robot World Cup Soccer Games and Conferences (1999)

....correlate them with the eld rectangular shape to determine the team postures. Other teams propose a vision based approach to self localization based on a single frontal camera, used to match a 3D geometric model of the eld with the border line segments and goal lines in the acquired image [16] [17]. RoboCup s Agilo team [16] proposes a single frontal camera to match a 3 D geometric model of the eld with the border lines and goals line segments in the acquired image. Only a partial eld view is used in this method. Iocchi and Nardi [17] also use a single frontal camera and match the lines ....

....line segments and goal lines in the acquired image [16] 17] RoboCup s Agilo team [16] proposes a single frontal camera to match a 3 D geometric model of the eld with the border lines and goals line segments in the acquired image. Only a partial eld view is used in this method. Iocchi and Nardi [17] also use a single frontal camera and match the lines with a eld model using the Hough Transform. Even though similar to the work on vision based self localization described in this paper, their approach considers lines detected locally (again due to a partial eld view) rather than 4 a global ....

L. Iocchi, D. Nardi, Self-localization in the robocup environment., in: Proc. of 16th IJCAI 99, The third International Workshop on RoboCup, 1999, pp. 116-120.

....with its teammates regarding team postures and ball location. The navigation system includes a guidance control algorithm which relies on odometry most of the time, but odometry is reset whenever the self localization algorithm runs. A similar method has been proposed by Iocchi and Nardi[15] for soccer robots too. Their method also matches the observed eld lines with a 2 D eld model in the Hough space. However, as only a single frontal camera is used, their approach considers lines detected locally, rather than a global eld view, and requires odometry to remove ambiguities. The ....

L. Iocchi and D. Nardi. Self-localization in the RoboCup environment. In M. Veloso, E. Pagello, and H. Kitano, editors, RoboCup-99: Robot Soccer World Cup III. Springer Verlag, Berlin, 2000.

....environment landmarks such as straight lines resulting from the intersection between the walls and the ground, as well as from a priori knowledge of the environment geometry. The correlation between the observed eld and its geometric model is made in Hough Transform space. Iocchi and Nardi [8] also use a single frontal camera and match the lines with a eld model using the Hough Transform. However, their approach considers lines detected locally, rather than a global eld view, and requires odometry to remove ambiguities. The paper is organized as follows: in Section 2, the proposed ....

L. Iocchi, D. Nardi, \Self-Localization in the RoboCup Environment", in Proc. of Sixteenth IJCAI 99, The III Int. Workshop on RoboCup, pp 116-120, 1999.

....Agilo team proposes a vision based approach to the self localization problem too. A single frontal camera is used to match a 3 D geometric model of the eld with the border lines and goals line segments in the acquired image. Only a partial eld view is used in this method. Iocchi and Nardi [6] also use a single frontal camera and match the lines with a eld model using the Hough Transform. Their approach considers lines detected locally, rather than a global eld view, and uses odometry to remove ambiguities. Several teams use a vision based omnidirectional hardware system similar to ....

L. Iocchi, D. Nardi. Self-Localization in the RoboCup Environment. Sixteenth IJCAI 99, The third International Workshop on RoboCup, pp 116-120,1999.

....Agilo team proposes a vision based approach to the self localization problem too. A single frontal camera is used to match a 3 D geometric model of the eld with the border lines and goals line segments in the acquired image. Only a partial eld view is used in this method. Iocchi and Nardi [6] also use a single frontal camera and match the lines with a eld model using the Hough Transform. Their approach considers lines detected locally, rather than a global eld view, and uses odometry to remove ambiguities. Several teams use a vision based omni directional hardware system similar to ....

L. Iocchi, D. Nardi. Self-Localization in the RoboCup Environment. Sixteenth IJCAI 99, The third International Workshop on RoboCup, pp 116-120,1999.

No context found.

L. Iocchi and D. Nardi. Self-localization in the RoboCup environment. In RoboCup-99: Robot Soccer World Cup III, pages 318-330, 1999. 24

....is presented, while the localization method in [6] is based on certainty grids and on the use of the Hough Transform only for extracting lines from these points. In this paper we present a method called Global Hough Localization, that extends the Hough Localization method described in [7] [8] in order to provide a solution for global localization. Global Hough Localization is based on matching a geometric reference map with a representation of range information acquired by the robot s sensors. We exploit the proper ties of the Hough Transform for recognizing lines from a sets of ....

....and tested this method in oce likeenvironments as well as in the RoboCup environmentby making use of vision based line extraction procedures performing as a range data sensor. II. Hough Localization In this section we recall the Hough Localization method for position tracking problem (see [7] [8] for a more detailed description) Hough Localization is based on a matching between aknown map of the environment and a local map built by the robot s sensors. The matching is performed between the Hough representation of both the reference map and the local map. Sensor data are thus transformed ....

[Article contains additional citation context not shown here]

L. Iocchi and D. Nardi, \Self-localization in the RoboCup environment," in ########### ##### ###### ##### ### ###, 1999, pp. 318-330.

No context found.

Luca Iocchi and Daniele Nardi. Self-localization in the robocup environment. In Manuela Veloso, Enrico Pagello, and Hiroaki Kitano, editors, RoboCup-99: Robot Soccer World Cup III, number 1856 in Lecture Notes in Artificial Intelligence, pages 318--330. Springer-Verlag, Berlin, 2000.

No context found.

L. Iocchi and D. Nardi. Self-Localization in the RoboCup Environment. In Third International RoboCup Symposium (Robot World Cup Soccer Games and Conferences), Lecture Notes in Computer Science. SpringerVerlag, 1999.

No context found.

Luca Iocchi and Daniele Nardi, "Self-Localization in the RoboCup Environment", Proc. Third International Workshop on RoboCup, p115

No context found.

IOCCHI, L., AND D., N. Self-localization in the robocup environment. In 3rd RoboCup Workshop, Springer- Verlag, in press. (1999).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC