S. Thrun. An online mapping algorithm for teams of mobile robots. Int. Journal of Robotics Research, 20(5):335--363, 2001.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....problem: Search in pose space: where a set of candidate vehicle locations is generated and analyzed looking for consistency between sensor measurements and the previous map. This idea, that can be used with raw sensor data, is the base of the Monte Carlo localization approach to SLAM [16]. Search in correspondence space: where sensor measurements are processed to obtain discrete features (points, lines, etc. that are matched against the features stored in the map. This is the approach used in all feature based approaches to SLAM [4] 6] 8] During continuous SLAM, the ....

S. Thrun. An online mapping algorithm for teams of mobile robots. Int. J. Robotics Research, 20(5):335--363, May 2001.

....its position. While there have been numerous published papers on the problem of CML in the past few years, there have been only a limited number of implementations of CML that have been operated in real time. To our knowledge, none of these implementations (with the possible exception of Thrun [1]) have used CML in realtime to actually control the motion of the robot. This paper describes a novel, real time implementation of CML running on a mobile robot in a dynamic indoor environment. The CML algorithm is actively used to navigate the robot. The precision and robustness of the CML ....

....the spectrum of representations is large, ranging from none at all [2] through grid based approaches [3] to probabilistic approaches. The later class have had the greatest success at achieving CML and can be further classified into feature based CML [4] 5] 6] 7] and data based estimation [1]. In this paper we adopt a feature based approach. In general proprioceptive sensor data is processed to estimate parameterized geometric representations of real world entities such as walls, corners or more complex compound objects. The entire environment is described by parameterizing 2D ....

S. Thrun, "An online mapping algorithm for teams of mobile robots," Int. J. Robotics Research, vol. 20, no. 5, pp. 335--363, May 2001.

....state of the world. Most implementations of the stochastic map use an extended Kalman filter [20] 8] 10] 3] The EKF will be used for illustration in this paper, but is not the only filter which could be used. For instance, an unscented filter [11] or sequential Monte Carlo algorithms [7] [21] could be chosen instead. For alternative approaches to CML that do not use a featurebased representation, see Thrun [21] Choset and Nagatani [5] and Kuipers [14] The basic stochastic mapping algorithm is summarized as follows [20] 8] A. Stochastic mapping algorithm 2: x(k k 1) f ( ....

....EKF will be used for illustration in this paper, but is not the only filter which could be used. For instance, an unscented filter [11] or sequential Monte Carlo algorithms [7] 21] could be chosen instead. For alternative approaches to CML that do not use a featurebased representation, see Thrun [21], Choset and Nagatani [5] and Kuipers [14] The basic stochastic mapping algorithm is summarized as follows [20] 8] A. Stochastic mapping algorithm 2: x(k k 1) f ( x(k 1 k 1) u(k) 3: Q u(k) noise covariance 4: F x 5: P = F x PF 6: z(k) sensors(x(k) ....

[Article contains additional citation context not shown here]

S. Thrun. An online mapping algorithm for teams of mobile robots. Int. J. Robotics Research, 20(5):335--363, May 2001.

....its position. While there have been numerous published papers on the problem of CML in the past few years, there have been only a limited number of implementations of CML that have been operated in real time. To our knowledge, none of these implementations (with the possible exception of Thrun [1]) have used CML in realtime to actually control the motion of the robot. This paper describes a novel, real time implementation of CML running on a mobile robot in a dynamic indoor environment. The CML algorithm is actively used to navigate the robot. The precision and robustness of the CML ....

....the spectrum of representations is large, ranging from none at all [2] through grid based approaches [3] to probabilistic approaches. The later class have had the greatest success at achieving CML and can be further classified into feature based CML [4] 5] 6] 7] and data based estimation [1]. In this paper we adopt a feature based approach. In general proprioceptive sensor data is processed to estimate parameterized geometric representations of real world entities such as walls, corners or more complex compound objects. The entire environment is described by parameterizing 2D ....

S. Thrun, "An online mapping algorithm for teams of mobile robots," Int. J. Robotics Research, vol. 20, no. 5, pp. 335--363, May 2001.

....different view point selection strategies for mobile robot exploration. 1 Introduction Generating maps is one of the fundamental tasks of mobile robots and many researchers have focused on the problem of how to represent the environment as well as how to acquire models using this representation [5, 9, 10, 14, 17] . The mapping problem itself has several aspects that have been studied intensively in the past. Some of the most important aspects are the localization of the vehicle during mapping, appropriate models of the environment and the sensors, as well as strategies for guiding the vehicle. In ....

S. Thrun. An online mapping algorithm for teams of mobile robots. International Journal of Robotics Research, 2001.

....RCM 3) the Office of Naval Research under grant N00014 970202, and by Draper Laboratories under contracts DL H 516617, DL H526716, and DL H 539054. matching, greatly eases the data association problem. This paper does not address the problems of large loop closing and global relocalization [7, 8]. An important objective of our work is to tie estimated scene structure to a common reference frame defined by the initial camera pose, as in the work in robotics known as simultaneous localization and mapping (SLAM) 9, 10, 11, 8] using laser range scanners [10, 12, 11, 13] Some vision ....

.... not address the problems of large loop closing and global relocalization [7, 8] An important objective of our work is to tie estimated scene structure to a common reference frame defined by the initial camera pose, as in the work in robotics known as simultaneous localization and mapping (SLAM) [9, 10, 11, 8], using laser range scanners [10, 12, 11, 13] Some vision researchers have pursued similar approaches for limited scenes [5, 14] 2. THE ALGORITHM VPs intersection line tracks clusters rotation updates translation updates edges 3D lines Fig. 1. Data Flow Graph. Figure 1 summarizes the ....

Thrun, S.: An online mapping algorithm for teams of mobile robots. Int. J. Robotics Research 20 (2001) 335--363

....a few tens of meters long. Figure 3 shows the accumulation of deadreckoning error during a longer duration traverse of about 500 meters in the MIT infinite corridor (shown in Figure 22) Most successful recent implementations of CML have either been performed with SICK laser scanner data [5, 6] or in environments that consist of isolated point objects [7, 8] However, there are many important applications of mobile robots where maps need to be built of complex environments, consisting of com Figure 1: Laser data for a short corridor experiment, referenced to the dead reckoning ....

....origin. posite features, from noisy sensor data. The goal of our work is to enable autonomous underwater vehicles to navigate autonomously using sonar. Current methods for data association in feature based CML are unable to cope with sonar because of its sparse and ambiguous nature. Thrun et al. [5] and Gutmann et al. 6] have developed implementations of CML using laser data that are capable of closing moderately sized loops in real time. In their work, the representation consists of raw sensor data referenced back to a complete trajectory of the vehicle. With this representation, they ....

S. Thrun. An online mapping algorithm for teams of mobile robots. Int. J. Robotics Research, 20(5):335-- 363, May 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. Int. Journal of Robotics Research, 20(5):335--363, 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. Int. Journal of Robotics Research, 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. International Journal of Robotics Research, 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. International Journal of Robotics Research, 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. International Journal of Robotics Research, 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. International Journal of Robotics Research, 2001.

No context found.

S. Thrun. An online mapping algorithm for teams of mobile robots. International Journal of Robotics Research, 2001.

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