nips nips2003 nips2003-21 knowledge-graph by maker-knowledge-mining

21 nips-2003-An Autonomous Robotic System for Mapping Abandoned Mines

Source: pdf

Author: David Ferguson, Aaron Morris, Dirk Hähnel, Christopher Baker, Zachary Omohundro, Carlos Reverte, Scott Thayer, Charles Whittaker, William Whittaker, Wolfram Burgard, Sebastian Thrun

Abstract: We present the software architecture of a robotic system for mapping abandoned mines. The software is capable of acquiring consistent 2D maps of large mines with many cycles, represented as Markov random £elds. 3D C-space maps are acquired from local 3D range scans, which are used to identify navigable paths using A* search. Our system has been deployed in three abandoned mines, two of which inaccessible to people, where it has acquired maps of unprecedented detail and accuracy. 1

Reference: text

Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 The software is capable of acquiring consistent 2D maps of large mines with many cycles, represented as Markov random £elds. [sent-13, score-0.418]

2 3D C-space maps are acquired from local 3D range scans, which are used to identify navigable paths using A* search. [sent-14, score-0.443]

3 Our system has been deployed in three abandoned mines, two of which inaccessible to people, where it has acquired maps of unprecedented detail and accuracy. [sent-15, score-0.666]

4 1 Introduction This paper describes the navigation software of a deployed robotic system for mapping subterranean spaces such as abandoned mines. [sent-16, score-0.503]

5 Subsidence of abandoned mines poses a major problem for society, as do ground water contaminations, mine £res, and so on. [sent-17, score-0.979]

6 Most abandoned mines are inaccessible to people, but some are accessible to robots. [sent-18, score-0.524]

7 When exploring and mapping unknown mines, it alternates short phases of motion guided by 2D range scans, with phases in which the vehicle rests to acquire 3D range scans. [sent-21, score-0.406]

8 An analysis of the 3D scans leads to a path that is then executed, using rapidly acquired 2D scans to determine the robot’s motion relative to the 3D map. [sent-22, score-0.616]

9 Acquiring consistent large-scale maps without external geo-referencing through GPS is largely considered an open research issue. [sent-24, score-0.206]

10 Our approach relies on ef£cient statistical techniques for generating such maps in real-time. [sent-25, score-0.174]

11 At the lowest level, we employ a fast scan matching algorithm for registering successive scans, thereby recovering robot odometry. [sent-26, score-0.568]

12 Groups of scans are then converted into local maps, using Markov random £eld representations (MRFs) to characterize the residual path uncertainty. [sent-27, score-0.315]

13 However, the brittleness of the ML approach is overcome by a “lazy” data association mechanism that can undo and redo past associations so as to maximize the overall map consistency. [sent-29, score-0.355]

14 To navigate, local 3D scans are mapped into 2 1 D terrain maps, by analyzing surface gradi2 ents and vertical clearance in the 3D scans. [sent-30, score-0.513]

15 The result is subsequently transformed into cost Figure 1: The Groundhog robot is a 1,500 pound custom-built vehicle equipped with onboard computing, laser range sensing, gas and sinkage sensors, and video recording equipment. [sent-31, score-0.588]

16 Its purpose is to explore and map abandoned mines. [sent-32, score-0.417]

17 functions expressed in the robot’s three-dimensional con£guration space, by convolving the 2 1 D terrain maps with kernels that describe the robot’s footprints in different orienta2 tions. [sent-33, score-0.432]

18 Some of the results reported here were obtained via manual control in mines accessible to people. [sent-36, score-0.242]

19 Others involved fully autonomous exploration, for which our robot operated fully self-guided for several hours beyond the reach of radio communication. [sent-37, score-0.38]

20 1 Generating Locally Consistent Maps As in [6, 9], we apply an incremental scan matching technique for registering scans, acquired using a forward-pointed laser range £nder while the vehicle is in motion. [sent-39, score-0.806]

21 This algorithm aligns scans by iteratively identifying nearby points in pairs of consecutive range scans, and then calculating the relative displacement and orientation of these scans by minimizing the quadratic distance of these pairs of points [2]. [sent-40, score-0.41]

22 This approach leads to the recovery of two quantities: locally consistent maps and an estimate of the robot’s motion. [sent-41, score-0.233]

23 It is well-understood [3, 6], however, that local scan matching is incapable of achieving globally consistent maps. [sent-42, score-0.384]

24 This is because of the residual error in scan matching, which accumulates over time. [sent-43, score-0.249]

25 The limitation is apparent in the map shown in Figure 2a, which is the result of applying local scan matching in a mine that is approximately 250 meters wide. [sent-44, score-1.089]

26 Our approach addresses this problem by explicitly representing the uncertainty in the map and the path using a Markov random £eld (MRF) [11]. [sent-45, score-0.225]

27 More speci£cally, the data acquired through every £ve meters of consecutive robot motion is mapped into a local map [3]. [sent-46, score-0.773]

28 The absolute location of orientation of the k-th map will be denoted by ξk = ( xk yk θk )T ; here x and y are the Cartesian coordinates and θ is the orientation. [sent-48, score-0.159]

29 Figure 3b shows the MRF for the data collected in the (a) (b) Figure 2: Mine map with incremental ML scan matching (left) and using our lazy data association approach (right). [sent-56, score-0.681]

30 2 Enforcing Global Consistency The key advantage of the MRF representation is that it encompasses the residual uncertainty in local scan matching. [sent-61, score-0.291]

31 This enables us to alter the shape of the map in accordance with global consistency constraints. [sent-62, score-0.159]

32 These constraints are obtained by matching local maps acquired at different points in time (e. [sent-63, score-0.45]

33 In particular, if the k-th map overlaps with some map j acquired at an earlier point in time, our approach localizes the robot relative to this map using once again local scan matching. [sent-66, score-1.137]

34 As a result, it recovers a relative constraint φ(ξk , ξj ) between the coordinates of non-adjacent maps ξk and ξj . [sent-67, score-0.174]

35 , the solution of the recursion (1) without the additional data association constraints). [sent-83, score-0.165]

36 (b) The Markov random £eld: Each node is the center of a local map, acquired when traversing the Bruceton Research Mine near Pittsburgh, PA. [sent-85, score-0.277]

37 ¯ ξj f (ξj , δk,j ) ¯ is the Jacobean of f (ξj , δk,j ) at ξj :   ¯ ¯ 1 0 −∆xk,j sin θk + ∆yk,j cos θk ¯ ¯ Fk,j x =  0 1 −∆xk,j cos θk − ∆yk,j sin θk  0 0 1 (5) The resulting negative log-likelihood is given by − log p(Ξ) ≈ const. [sent-86, score-0.168]

38 Empirically, we £nd that this inversion can be performed very ef£ciently using an inversion algorithm described in [15]; it only requires a few seconds for matrices composed of hundreds of local map positions (and it appears to be numerically more stable than the solution in [11, 6]). [sent-91, score-0.201]

39 3 Lazy Data Association Search Unfortunately, the approach described thus far leads only to a consistent map when the additional constraints φ(ξk , ξj ) obtained after loop closure are correct. [sent-95, score-0.358]

40 These constraints amount to a maximum likelihood solution for the challenging data association problem that arises when closing a loop. [sent-96, score-0.229]

41 Figure 4a depicts such a situation, obtained when operating our vehicle in a large abandoned mine. [sent-98, score-0.492]

42 As long as the correct such closure is in the set of surviving particle £lters, the correct map can be recovered. [sent-102, score-0.255]

43 In the context of our present system, this approach suffers from two disadvantages: it is computationally expensive due to its proactive nature, and it provides no mechanism for recovery should the correct loop closure not be represented in the particle set. [sent-103, score-0.237]

44 However, it also provides a mechanism to undo and redo past data association decisions. [sent-106, score-0.196]

45 If this likelihood falls below a threshold, data association constraints are recursively undone and replaced by other constraints of decreasing likelihood (including the possibility of not generating a constraint at all). [sent-108, score-0.218]

46 The search terminates if the likelihood of the most recent measurement (a) (b) con¤ict ¨ % ¨ map after adjustment ' start Figure 4: Example of our lazy data association technique: When closing a large loop, the robot £rst erroneously assumes the existence of a second, parallel hallway. [sent-109, score-0.742]

47 However, this model leads to a gross inconsistency as the robot encounters a corridor at a right angle. [sent-110, score-0.361]

48 At this point, our approach recursively searches for improved data association decisions, arriving at the map shown on the right. [sent-111, score-0.309]

49 The left panel shows the ML association after traversing a large loop inside a mine: At £rst, it appears that the existence of two adjacent corridors is more likely than a single one, according to the estimated robot motion. [sent-115, score-0.56]

50 However, as the robot approaches a turn, a noticeable inconsistency is detected. [sent-116, score-0.259]

51 As a result, our data association mechanism recursively removes past data association constraints back to the most recent loop closure, and then “tries” the second most likely hypothesis. [sent-118, score-0.383]

52 The backtracking requires a fraction of a second, and with high likelihood leads to a globally consistent map and, as a side-effect, to an improved estimate of the map coordinates Ξ. [sent-120, score-0.415]

53 3 Autonomous Navigation 2D maps are suf£cient for localizing robots inside mines; however, they are insuf£cient to navigate a robot due to the rugged nature of abandoned mines. [sent-122, score-0.869]

54 Our approach to navigation is based on 3D maps, acquired in periodic intervals while the vehicle suspends motion to scan its environment. [sent-123, score-0.681]

55 A typical 3D scan is shown in Figure 5a; others are shown in Figure 7. [sent-124, score-0.218]

56 1 1 2 2 D Terrain Maps In a £rst processing step, the robot projects local 3D maps onto 2 1 D terrain maps, such as 2 the one shown in Figure 5b. [sent-126, score-0.733]

57 The gray-level in this map illustrates the degree at which the map is traversable: the brighter a 2D location, the better suited it is for navigation. [sent-127, score-0.355]

58 The terrain map is obtained by analyzing all measurements x, y, z in the 3D scan (where z is the vertical dimension). [sent-128, score-0.635]

59 It then searches for the largest z value in this region whose distance to z does not exceed the vehicle height (plus a safety margin); this value will be called z . [sent-130, score-0.263]

60 The difference z − z is the navigational coef£cient: ¯ ¯ it loosely corresponds to the ruggedness of the terrain under the height of the robot. [sent-131, score-0.258]

61 For safety reasons, multiple regions {xmin ; xmax } × {ymin ; ymax } overlap when building the terrain map. [sent-133, score-0.397]

62 The terrain map is subsequently convolved with a narrow radial kernel that serves as a repellent potential £eld, to keep the robot clear of obstacles. [sent-134, score-0.676]

63 2 Con£guration Space Maps The terrain map is used to construct a collection of maps that describe the robot’s con£guration space, or C-space [10]. [sent-136, score-0.591]

64 The C-space is the three-dimensional space of poses that (a) (b) (c) Figure 5: (a) A local 3D model of the mine corridor, obtained by a scanning laser range £nder. [sent-137, score-0.674]

65 (b) 1 The corresponding 2 2 D terrain map extracted from this 3D snapshot: the brighter a location, the 1 easier it is to navigate. [sent-138, score-0.454]

66 (c) Kernels for generating directional C-space maps from the 2 2 D terrain map. [sent-139, score-0.432]

67 Planning in these C-space maps ensures that the terrain under the tires is maximally navigable. [sent-141, score-0.469]

68 the vehicle can assume; it comprises the x-y location along with the vehicle’s orientation θ. [sent-142, score-0.234]

69 The C-space maps are obtained by convolving the terrain map with oriented kernels that describe the robot’s footprint. [sent-143, score-0.591]

70 The intuition of using such a kernel is as follows: Abandoned mines often possess railroad tracks, and while it is perfectly acceptable to navigate with a track between the wheels, traversing or riding these tracks causes unnecessary damage to the tires and will increase the energy consumption. [sent-145, score-0.331]

71 The result of this transformation is a collection of C-space maps, each of which applies to a different vehicle orientation. [sent-146, score-0.234]

72 The A* search is initiated with an array of goal points, which places the highest value at locations at maximum distance straight down a mine corridor. [sent-149, score-0.496]

73 5 meters), the robot decides that the hallway is not navigable and initiates a high-level decision to turn around. [sent-152, score-0.341]

74 4 Results The approach was tested in multiple experiments, some of which were remotely operated while in others the robot operated autonomously, outside the reach of radio communication. [sent-154, score-0.39]

75 Figure 6b shows a picture of the tethered and remotely controlled vehicle inside this mine, which has not been entered by people for many decades. [sent-156, score-0.343]

76 Its partially ¤ooded nature prevented an entry into the mine for more than approximately 40 meters. [sent-157, score-0.496]

77 Maps acquired in this mine are shown in Figure 9. [sent-158, score-0.637]

78 On May 30, 2003, Groundhog successfully explored an abandoned mine using the fully autonomous mode. [sent-159, score-0.811]

79 The mine, known as the Mathies Mine near Pittsburgh, is part of a large mine system near Courtney, PA. [sent-160, score-0.574]

80 Existing maps for this mine are highly inaccurate, and the conditions inside the mine were unknown to us. [sent-161, score-1.213]

81 Figure 6a shows the robot as it enters the mine, and Figure 7a depicts a typical 3D scan acquired in the entrance area. [sent-162, score-0.618]

82 (a) (b) Figure 6: (a) The vehicle as it enters the Mathies Mine on May 30, 2003. [sent-163, score-0.234]

83 It autonomously descended 308 meters into the mine before making the correct decision to turn around due to a blockage inside the mine. [sent-164, score-0.702]

84 (b) The vehicle, as it negotiates acidic mud under manual remote control approximately 30 meters into the Florence Mine near Burgettstown, PA. [sent-165, score-0.217]

85 (a) (b) Figure 7: 3D local maps: (a) a typical corridor map that is highly navigable. [sent-166, score-0.274]

86 (b) a map of a broken ceiling bar that renders the corridor segment unnavigable. [sent-167, score-0.29]

87 This obstacle was encountered 308 meters into the abandoned Mathies Mine. [sent-168, score-0.411]

88 Figure 8: Fraction of the 2D mine map of the Mathies Mine, autonomously explored by the Groundhog vehicle. [sent-169, score-0.699]

89 Also shown is the path of the robot and the locations at which it chose to take 3D scans. [sent-170, score-0.325]

90 (a) (b) (c) Figure 9: (a) A small 2D map acquired by Groundhog in the Florence Mine near Burgettstown, PA. [sent-172, score-0.339]

91 (c) Image acquired by Groundhog inside the Mathies Mine (a dry mine). [sent-175, score-0.188]

92 After successfully descending 308 meters into the Mathies Mine, negotiating some rough terrain along the way, the robot encountered a broken ceiling beam that draped diagonally across the robot’s path. [sent-176, score-0.69]

93 The corresponding 3D scan is shown in Figure 7b: it shows rubble on the ground, along with the ceiling bar and two ceiling cables dragged down by the bar. [sent-177, score-0.334]

94 The robot’s A* motion planner failed to identify a navigable path, and the robot made the appropriate decision to retreat. [sent-178, score-0.371]

95 Figure 8 shows the corresponding 2D map; the entire map is 308 meters long, but here we only show the £nal section, along with the path and the location at which the robot stop to take a 3D scan. [sent-179, score-0.599]

96 An image acquired in this mine is depicted in Figure 9c. [sent-180, score-0.637]

97 5 Conclusion We have described the software architecture of a deployed system for robotic mine mapping. [sent-181, score-0.62]

98 The system has been tested under extreme conditions, and generated accurate maps of abandoned mines inaccessible to people. [sent-183, score-0.671]

99 A case study in robotic mapping of abandoned mines. [sent-194, score-0.369]

100 A highly ef£cient FastSLAM algorithm for generating cyclic maps of large-scale environments from raw laser range measurements. [sent-238, score-0.269]

similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('mine', 0.496), ('robot', 0.259), ('abandoned', 0.258), ('terrain', 0.258), ('vehicle', 0.234), ('scan', 0.218), ('mines', 0.184), ('scans', 0.176), ('maps', 0.174), ('map', 0.159), ('mrf', 0.146), ('acquired', 0.141), ('association', 0.122), ('meters', 0.115), ('groundhog', 0.111), ('mathies', 0.111), ('lazy', 0.096), ('loop', 0.077), ('closing', 0.073), ('corridor', 0.073), ('path', 0.066), ('laser', 0.064), ('matching', 0.059), ('ceiling', 0.058), ('robotic', 0.058), ('motion', 0.057), ('autonomous', 0.057), ('closure', 0.056), ('ahnel', 0.055), ('bruceton', 0.055), ('burgettstown', 0.055), ('inaccessible', 0.055), ('navigable', 0.055), ('navigate', 0.055), ('traversing', 0.055), ('xmax', 0.055), ('ymax', 0.055), ('ymin', 0.055), ('mapping', 0.053), ('burgard', 0.048), ('florence', 0.048), ('inconsistencies', 0.048), ('inside', 0.047), ('ml', 0.045), ('autonomously', 0.044), ('cycles', 0.043), ('recursion', 0.043), ('local', 0.042), ('cos', 0.042), ('sin', 0.042), ('poses', 0.041), ('particle', 0.04), ('robots', 0.039), ('near', 0.039), ('deployed', 0.038), ('obstacle', 0.038), ('brighter', 0.037), ('clearance', 0.037), ('fastslam', 0.037), ('montemerlo', 0.037), ('ooded', 0.037), ('proactive', 0.037), ('redo', 0.037), ('rugged', 0.037), ('subterranean', 0.037), ('tires', 0.037), ('undo', 0.037), ('acknowledge', 0.037), ('exploration', 0.036), ('operated', 0.035), ('guration', 0.034), ('constraints', 0.034), ('globally', 0.033), ('measurement', 0.033), ('consistent', 0.032), ('backtracking', 0.032), ('freiburg', 0.032), ('mud', 0.032), ('registering', 0.032), ('remotely', 0.032), ('xmin', 0.032), ('range', 0.031), ('residual', 0.031), ('navigation', 0.031), ('localization', 0.031), ('manual', 0.031), ('people', 0.03), ('radio', 0.029), ('mrfs', 0.029), ('safety', 0.029), ('fox', 0.029), ('gross', 0.029), ('recursively', 0.028), ('loopy', 0.028), ('software', 0.028), ('accessible', 0.027), ('displacement', 0.027), ('hallway', 0.027), ('recovery', 0.027), ('incremental', 0.027)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.99999994 21 nips-2003-An Autonomous Robotic System for Mapping Abandoned Mines

Author: David Ferguson, Aaron Morris, Dirk Hähnel, Christopher Baker, Zachary Omohundro, Carlos Reverte, Scott Thayer, Charles Whittaker, William Whittaker, Wolfram Burgard, Sebastian Thrun

2 0.13299403 36 nips-2003-Auction Mechanism Design for Multi-Robot Coordination

Author: Curt Bererton, Geoffrey J. Gordon, Sebastian Thrun

Abstract: The design of cooperative multi-robot systems is a highly active research area in robotics. Two lines of research in particular have generated interest: the solution of large, weakly coupled MDPs, and the design and implementation of market architectures. We propose a new algorithm which joins together these two lines of research. For a class of coupled MDPs, our algorithm automatically designs a market architecture which causes a decentralized multi-robot system to converge to a consistent policy. We can show that this policy is the same as the one which would be produced by a particular centralized planning algorithm. We demonstrate the new algorithm on three simulation examples: multi-robot towing, multi-robot path planning with a limited fuel resource, and coordinating behaviors in a game of paint ball. 1

3 0.085882455 54 nips-2003-Discriminative Fields for Modeling Spatial Dependencies in Natural Images

Author: Sanjiv Kumar, Martial Hebert

Abstract: In this paper we present Discriminative Random Fields (DRF), a discriminative framework for the classiﬁcation of natural image regions by incorporating neighborhood spatial dependencies in the labels as well as the observed data. The proposed model exploits local discriminative models and allows to relax the assumption of conditional independence of the observed data given the labels, commonly used in the Markov Random Field (MRF) framework. The parameters of the DRF model are learned using penalized maximum pseudo-likelihood method. Furthermore, the form of the DRF model allows the MAP inference for binary classiﬁcation problems using the graph min-cut algorithms. The performance of the model was veriﬁed on the synthetic as well as the real-world images. The DRF model outperforms the MRF model in the experiments. 1

4 0.076950118 33 nips-2003-Approximate Planning in POMDPs with Macro-Actions

Author: Georgios Theocharous, Leslie P. Kaelbling

Abstract: Recent research has demonstrated that useful POMDP solutions do not require consideration of the entire belief space. We extend this idea with the notion of temporal abstraction. We present and explore a new reinforcement learning algorithm over grid-points in belief space, which uses macro-actions and Monte Carlo updates of the Q-values. We apply the algorithm to a large scale robot navigation task and demonstrate that with temporal abstraction we can consider an even smaller part of the belief space, we can learn POMDP policies faster, and we can do information gathering more efﬁciently.

5 0.070607573 2 nips-2003-ARA*: Anytime A* with Provable Bounds on Sub-Optimality

Author: Maxim Likhachev, Geoffrey J. Gordon, Sebastian Thrun

Abstract: In real world planning problems, time for deliberation is often limited. Anytime planners are well suited for these problems: they ﬁnd a feasible solution quickly and then continually work on improving it until time runs out. In this paper we propose an anytime heuristic search, ARA*, which tunes its performance bound based on available search time. It starts by ﬁnding a suboptimal solution quickly using a loose bound, then tightens the bound progressively as time allows. Given enough time it ﬁnds a provably optimal solution. While improving its bound, ARA* reuses previous search efforts and, as a result, is signiﬁcantly more efﬁcient than other anytime search methods. In addition to our theoretical analysis, we demonstrate the practical utility of ARA* with experiments on a simulated robot kinematic arm and a dynamic path planning problem for an outdoor rover. 1

6 0.070352688 43 nips-2003-Bounded Invariance and the Formation of Place Fields

7 0.055305276 8 nips-2003-A Holistic Approach to Compositional Semantics: a connectionist model and robot experiments

8 0.04900641 29 nips-2003-Applying Metric-Trees to Belief-Point POMDPs

9 0.044884283 28 nips-2003-Application of SVMs for Colour Classification and Collision Detection with AIBO Robots

10 0.04391522 37 nips-2003-Automatic Annotation of Everyday Movements

11 0.042029317 160 nips-2003-Prediction on Spike Data Using Kernel Algorithms

12 0.040977184 158 nips-2003-Policy Search by Dynamic Programming

13 0.037974458 112 nips-2003-Learning to Find Pre-Images

14 0.037709948 167 nips-2003-Robustness in Markov Decision Problems with Uncertain Transition Matrices

15 0.036778316 35 nips-2003-Attractive People: Assembling Loose-Limbed Models using Non-parametric Belief Propagation

16 0.036245156 32 nips-2003-Approximate Expectation Maximization

17 0.034818355 7 nips-2003-A Functional Architecture for Motion Pattern Processing in MSTd

18 0.03395617 69 nips-2003-Factorization with Uncertainty and Missing Data: Exploiting Temporal Coherence

19 0.03384082 115 nips-2003-Linear Dependent Dimensionality Reduction

20 0.033554666 16 nips-2003-A Recurrent Model of Orientation Maps with Simple and Complex Cells

similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, -0.126), (1, 0.046), (2, 0.009), (3, 0.022), (4, -0.078), (5, -0.061), (6, 0.001), (7, -0.034), (8, 0.06), (9, -0.021), (10, 0.027), (11, -0.015), (12, 0.002), (13, 0.097), (14, 0.03), (15, 0.04), (16, 0.052), (17, 0.002), (18, -0.015), (19, 0.05), (20, 0.036), (21, -0.057), (22, -0.062), (23, -0.022), (24, 0.085), (25, -0.05), (26, -0.151), (27, 0.06), (28, -0.02), (29, 0.139), (30, 0.078), (31, -0.134), (32, 0.008), (33, 0.025), (34, 0.229), (35, -0.116), (36, -0.135), (37, 0.008), (38, 0.018), (39, 0.145), (40, -0.047), (41, -0.055), (42, 0.028), (43, 0.048), (44, 0.04), (45, -0.087), (46, -0.165), (47, 0.135), (48, 0.171), (49, 0.09)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.96620655 21 nips-2003-An Autonomous Robotic System for Mapping Abandoned Mines

Author: David Ferguson, Aaron Morris, Dirk Hähnel, Christopher Baker, Zachary Omohundro, Carlos Reverte, Scott Thayer, Charles Whittaker, William Whittaker, Wolfram Burgard, Sebastian Thrun

2 0.7549296 2 nips-2003-ARA*: Anytime A* with Provable Bounds on Sub-Optimality

Author: Maxim Likhachev, Geoffrey J. Gordon, Sebastian Thrun

3 0.68895584 36 nips-2003-Auction Mechanism Design for Multi-Robot Coordination

Author: Curt Bererton, Geoffrey J. Gordon, Sebastian Thrun

4 0.50217873 8 nips-2003-A Holistic Approach to Compositional Semantics: a connectionist model and robot experiments

Author: Yuuya Sugita, Jun Tani

Abstract: We present a novel connectionist model for acquiring the semantics of a simple language through the behavioral experiences of a real robot. We focus on the “compositionality” of semantics, a fundamental characteristic of human language, which is the ability to understand the meaning of a sentence as a combination of the meanings of words. We also pay much attention to the “embodiment” of a robot, which means that the robot should acquire semantics which matches its body, or sensory-motor system. The essential claim is that an embodied compositional semantic representation can be self-organized from generalized correspondences between sentences and behavioral patterns. This claim is examined and conﬁrmed through simple experiments in which a robot generates corresponding behaviors from unlearned sentences by analogy with the correspondences between learned sentences and behaviors. 1

5 0.38734224 43 nips-2003-Bounded Invariance and the Formation of Place Fields

Author: Reto Wyss, Paul F. Verschure

Abstract: One current explanation of the view independent representation of space by the place-cells of the hippocampus is that they arise out of the summation of view dependent Gaussians. This proposal assumes that visual representations show bounded invariance. Here we investigate whether a recently proposed visual encoding scheme called the temporal population code can provide such representations. Our analysis is based on the behavior of a simulated robot in a virtual environment containing speciﬁc visual cues. Our results show that the temporal population code provides a representational substrate that can naturally account for the formation of place ﬁelds. 1

6 0.38138658 6 nips-2003-A Fast Multi-Resolution Method for Detection of Significant Spatial Disease Clusters

7 0.37464368 54 nips-2003-Discriminative Fields for Modeling Spatial Dependencies in Natural Images

8 0.32057381 76 nips-2003-GPPS: A Gaussian Process Positioning System for Cellular Networks

9 0.31667235 28 nips-2003-Application of SVMs for Colour Classification and Collision Detection with AIBO Robots

10 0.26243225 33 nips-2003-Approximate Planning in POMDPs with Macro-Actions

11 0.24089086 44 nips-2003-Can We Learn to Beat the Best Stock

12 0.23670503 196 nips-2003-Wormholes Improve Contrastive Divergence

13 0.22704223 61 nips-2003-Entrainment of Silicon Central Pattern Generators for Legged Locomotory Control

14 0.22578304 138 nips-2003-Non-linear CCA and PCA by Alignment of Local Models

15 0.22415383 37 nips-2003-Automatic Annotation of Everyday Movements

16 0.19580705 29 nips-2003-Applying Metric-Trees to Belief-Point POMDPs

17 0.19250955 25 nips-2003-An MCMC-Based Method of Comparing Connectionist Models in Cognitive Science

18 0.19002448 7 nips-2003-A Functional Architecture for Motion Pattern Processing in MSTd

19 0.18977298 168 nips-2003-Salient Boundary Detection using Ratio Contour

20 0.18874978 191 nips-2003-Unsupervised Context Sensitive Language Acquisition from a Large Corpus

similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(0, 0.047), (11, 0.03), (29, 0.019), (30, 0.016), (35, 0.044), (49, 0.01), (53, 0.062), (69, 0.02), (71, 0.054), (76, 0.051), (82, 0.411), (85, 0.066), (91, 0.074), (99, 0.017)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.81928927 21 nips-2003-An Autonomous Robotic System for Mapping Abandoned Mines

Author: David Ferguson, Aaron Morris, Dirk Hähnel, Christopher Baker, Zachary Omohundro, Carlos Reverte, Scott Thayer, Charles Whittaker, William Whittaker, Wolfram Burgard, Sebastian Thrun

2 0.72381705 181 nips-2003-Statistical Debugging of Sampled Programs

Author: Alice X. Zheng, Michael I. Jordan, Ben Liblit, Alex Aiken

Abstract: We present a novel strategy for automatically debugging programs given sampled data from thousands of actual user runs. Our goal is to pinpoint those features that are most correlated with crashes. This is accomplished by maximizing an appropriately deﬁned utility function. It has analogies with intuitive debugging heuristics, and, as we demonstrate, is able to deal with various types of bugs that occur in real programs. 1

3 0.65346187 139 nips-2003-Nonlinear Filtering of Electron Micrographs by Means of Support Vector Regression

Author: Roland Vollgraf, Michael Scholz, Ian A. Meinertzhagen, Klaus Obermayer

Abstract: Nonlinear ﬁltering can solve very complex problems, but typically involve very time consuming calculations. Here we show that for ﬁlters that are constructed as a RBF network with Gaussian basis functions, a decomposition into linear ﬁlters exists, which can be computed eﬃciently in the frequency domain, yielding dramatic improvement in speed. We present an application of this idea to image processing. In electron micrograph images of photoreceptor terminals of the fruit ﬂy, Drosophila, synaptic vesicles containing neurotransmitter should be detected and labeled automatically. We use hand labels, provided by human experts, to learn a RBF ﬁlter using Support Vector Regression with Gaussian kernels. We will show that the resulting nonlinear ﬁlter solves the task to a degree of accuracy, which is close to what can be achieved by human experts. This allows the very time consuming task of data evaluation to be done eﬃciently. 1

4 0.36453116 38 nips-2003-Autonomous Helicopter Flight via Reinforcement Learning

Author: H. J. Kim, Michael I. Jordan, Shankar Sastry, Andrew Y. Ng

Abstract: Autonomous helicopter ﬂight represents a challenging control problem, with complex, noisy, dynamics. In this paper, we describe a successful application of reinforcement learning to autonomous helicopter ﬂight. We ﬁrst ﬁt a stochastic, nonlinear model of the helicopter dynamics. We then use the model to learn to hover in place, and to ﬂy a number of maneuvers taken from an RC helicopter competition.

5 0.36187351 57 nips-2003-Dynamical Modeling with Kernels for Nonlinear Time Series Prediction

Author: Liva Ralaivola, Florence D'alché-buc

Abstract: We consider the question of predicting nonlinear time series. Kernel Dynamical Modeling (KDM), a new method based on kernels, is proposed as an extension to linear dynamical models. The kernel trick is used twice: ﬁrst, to learn the parameters of the model, and second, to compute preimages of the time series predicted in the feature space by means of Support Vector Regression. Our model shows strong connection with the classic Kalman Filter model, with the kernel feature space as hidden state space. Kernel Dynamical Modeling is tested against two benchmark time series and achieves high quality predictions. 1

6 0.34389108 78 nips-2003-Gaussian Processes in Reinforcement Learning

7 0.34088799 109 nips-2003-Learning a Rare Event Detection Cascade by Direct Feature Selection

8 0.34079322 113 nips-2003-Learning with Local and Global Consistency

9 0.34066138 189 nips-2003-Tree-structured Approximations by Expectation Propagation

10 0.34041226 54 nips-2003-Discriminative Fields for Modeling Spatial Dependencies in Natural Images

11 0.3383961 116 nips-2003-Linear Program Approximations for Factored Continuous-State Markov Decision Processes

12 0.33776888 20 nips-2003-All learning is Local: Multi-agent Learning in Global Reward Games

13 0.33774531 50 nips-2003-Denoising and Untangling Graphs Using Degree Priors

14 0.33735728 158 nips-2003-Policy Search by Dynamic Programming

15 0.33698213 101 nips-2003-Large Margin Classifiers: Convex Loss, Low Noise, and Convergence Rates

16 0.33668637 112 nips-2003-Learning to Find Pre-Images

17 0.33655006 103 nips-2003-Learning Bounds for a Generalized Family of Bayesian Posterior Distributions

18 0.33582112 172 nips-2003-Semi-Supervised Learning with Trees

19 0.33513838 3 nips-2003-AUC Optimization vs. Error Rate Minimization

20 0.33498597 147 nips-2003-Online Learning via Global Feedback for Phrase Recognition